NCHRP - 219 Report - Hamburg Wheel-Track Test Equipment

This PDF is available at http://nap.nationalacademies.
org/21931
Hamburg Wheel-Track Test Equipment

Requirements and Improvements to AASHTO
T 324 (2016)
DETAILS
88 pages | 8.5 x 11 | PAPERBACK
ISBN 978-0-309-43109-5 | DOI 10.17226/21931
CONTRIBUTORS
Mohammad, Louay N.; Elseifi, Mostafa A.; Raghavendra, Amar; and Ye Mengqiu
BUY THIS BOOK
FIND RELATED TITLES SUGGESTED CITATION

National Academies of Sciences, Engineering, and Medicine. 2016. Hamburg
Wheel-Track Test Equipment Requirements and Improvements to AASHTO T 324.
Washington, DC: The National Academies Press. https://doi.org/10.17226/21931.
Visit the National Academies Press at nap.edu and login or register to get:
– Access to free PDF downloads of thousands of publications
– 10% off the price of print publications
– Email or social media notiﬁcations of new titles related to your interests
– Special offers and discounts
All downloadable National Academies titles are free to be used for personal and/or non-commercial
academic use. Users may also freely post links to our titles on this website; non-commercial academic
users are encouraged to link to the version on this website rather than distribute a downloaded PDF
to ensure that all users are accessing the latest authoritative version of the work. All other uses require
written permission. (Request Permission)
This PDF is protected by copyright and owned by the National Academy of Sciences; unless otherwise
indicated, the National Academy of Sciences retains copyright to all materials in this PDF with all rights
reserved.
Hamburg Wheel-Track Test Equipment Requirements and Improvements to AASHTO T 324
NCHRP
Web-Only Document 219:
Hamburg Wheel-Track Test

Equipment Requirements
and Improvements to
AASHTO T 324
Louay N. Mohammad
Mostafa A. Elseifi
Amar Raghavendra
Mengqiu Ye
Louisiana Transportation Research Center
Louisiana State University
Baton Rouge, LA
Contractor’s Final Report for NCHRP Project 20-07/Task 361

Submitted September 2015
Copyright National Academy of Sciences. All rights reserved.

ACKNOWLEDGMENT
This work was sponsored by the American Association of State

Highway and Transportation Officials (AASHTO), in cooperation with
the Federal Highway Administration, and was conducted in the
National Cooperative Highway Research Program (NCHRP), which is
administered by the Transportation Research Board (TRB) of the
National Academies of Sciences, Engineering, and Medicine.
COPYRIGHT INFORMATION
Authors herein are responsible for the authenticity of their materials

and for obtaining written permissions from publishers or persons who
own the copyright to any previously published or copyrighted material
used herein.
Cooperative Research Programs (CRP) grants permission to reproduce

material in this publication for classroom and not-for-profit purposes.
Permission is given with the understanding that none of the material
will be used to imply TRB, AASHTO, FAA, FHWA, FMCSA, FRA,
FTA, Office of the Assistant Secretary for Research and Technology,
PHMSA, or TDC endorsement of a particular product, method, or
practice. It is expected that those reproducing the material in this
document for educational and not-for-profit uses will give appropriate
acknowledgment of the source of any reprinted or reproduced material.
For other uses of the material, request permission from CRP.
DISCLAIMER
The opinions and conclusions expressed or implied in this report are

those of the researchers who performed the research. They are not
necessarily those of the Transportation Research Board; the National
Academies of Sciences, Engineering, and Medicine; or the program
sponsors.
The information contained in this document was taken directly from the
submission of the author(s). This material has not been edited by TRB.

The National Academy of Sciences was established in 1863 by an Act of Congress, signed by President Lincoln, as a private, non-
governmental institution to advise the nation on issues related to science and technology. Members are elected by their peers for
outstanding contributions to research. Dr. Ralph J. Cicerone is president.
The National Academy of Engineering was established in 1964 under the charter of the National Academy of Sciences to bring the
practices of engineering to advising the nation. Members are elected by their peers for extraordinary contributions to engineering.
Dr. C. D. Mote, Jr., is president.
The National Academy of Medicine (formerly the Institute of Medicine) was established in 1970 under the charter of the National
Academy of Sciences to advise the nation on medical and health issues. Members are elected by their peers for distinguished contributions
to medicine and health. Dr. Victor J. Dzau is president.
The three Academies work together as the National Academies of Sciences, Engineering, and Medicine to provide independent,
objective analysis and advice to the nation and conduct other activities to solve complex problems and inform public policy decisions.
The Academies also encourage education and research, recognize outstanding contributions to knowledge, and increase public
understanding in matters of science, engineering, and medicine.
Learn more about the National Academies of Sciences, Engineering, and Medicine at www.national-academies.org.
The Transportation Research Board is one of seven major programs of the National Academies of Sciences, Engineering, and Medicine.
The mission of the Transportation Research Board is to increase the benefits that transportation contributes to society by providing
leadership in transportation innovation and progress through research and information exchange, conducted within a setting that is
objective, interdisciplinary, and multimodal. The Board’s varied committees, task forces, and panels annually engage about 7,000
engineers, scientists, and other transportation researchers and practitioners from the public and private sectors and academia, all of
whom contribute their expertise in the public interest. The program is supported by state transportation departments, federal agencies
including the component administrations of the U.S. Department of Transportation, and other organizations and individuals interested
in the development of transportation.
Learn more about the Transportation Research Board at www.TRB.org.

Table of Contents
1. Summary ........................................................................................................................................ 3
2. Introduction and Research Approach ............................................................................................. 5
2.1. Problem Statement.................................................................................................................. 5
2.2. Research Objective ................................................................................................................. 6
2.3. Research Approach................................................................................................................. 6
2.3.1 Task 1 – Available Hamburg Test Equipment Specifications ........................................ 6
2.3.2 Task 2 – Engineering Desk Analysis of Existing Hamburg Test Systems ..................... 6
2.3.3 Task 3 – Propose Revisions to AASHTO T 324 ............................................................ 7
2.3.4 Task 4 – A Framework for Future Laboratory Evaluation ............................................. 7
2.3.5 Task 5 – Prepare and Submit Final Report ..................................................................... 7
3. Methodology and Experimental Program ...................................................................................... 8
3.1. Review of Test Equipment Specifications.............................................................................. 8
3.2. Nationwide Survey ................................................................................................................. 8
3.3. Experimental Program ............................................................................................................ 8
3.3.1 Wheel Position Waveform, Frequency, and Maximum Speed....................................... 9
3.3.2 Impression Measurement System ................................................................................. 12
3.3.3 Temperature Control System ........................................................................................ 16
3.3.4 Wheel Dimensions........................................................................................................ 17
3.3.5 Wheel Loads ................................................................................................................. 18
3.3.6 Specimen and Track Length ......................................................................................... 18
3.3.7 Free Circulating Water on Mounting System ............................................................... 19
3.3.8 Data Collection and Reporting ..................................................................................... 20
4. Results and Analysis .................................................................................................................... 21
4.1. Review of Test Equipment Specifications............................................................................ 21
4.1.1 Loading Mechanism ..................................................................................................... 24
4.1.2 Wheel Speed ................................................................................................................. 24
4.1.3 Temperature Measurement and Control System .......................................................... 25
4.1.4 Impression Measurement System ................................................................................. 25
4.1.5 Specimen Length and Track Length ............................................................................. 25
4.1.6 Data Collection and Reporting ..................................................................................... 25
4.2. Findings of the Nationwide Survey ...................................................................................... 26

4.2.1 Type of HWTs Used ..................................................................................................... 27

4.2.2 Test Methods Used ....................................................................................................... 27
4.2.3 Test Temperature .......................................................................................................... 28
4.2.4 Mix Acceptance Criteria............................................................................................... 29
4.2.5 Reporting of Test Results ............................................................................................. 30
4.3. Wheel Position Waveform, Frequency, and Maximum Speed............................................. 30
4.3.1 Test Results .................................................................................................................. 30
4.4. Experiment I – Wheel Dimensions....................................................................................... 36
4.5. Experiment II – Wheel Loads............................................................................................... 38
4.6. Experiment III – Free Circulating Water on Mounting System ........................................... 39
4.7. Experiment IV – Temperature Measurement and Control System ...................................... 39
4.7.1 Evaluation at 50°C........................................................................................................ 40
4.7.2 Evaluation at 25°C........................................................................................................ 45
4.7.3 Evaluation at 70°C........................................................................................................ 48
4.8. Experiment V – Impression Measurement System .............................................................. 51
4.8.1 Test Results .................................................................................................................. 52
4.9. Data Collection and Reporting ............................................................................................. 56
4.9.1 Test Materials ............................................................................................................... 58
4.9.2 Results and Analysis..................................................................................................... 58
4.10. Proposed Modifications to AASHTO T324 Specifications.............................................. 63
4.11. Laboratory Experimental Plan for Validation of Proposed Changes ............................... 65
5. Conclusions and Recommendations............................................................................................. 66
5.1. Recommendations ................................................................................................................ 67
6. References .................................................................................................................................... 68
7. Appendix A .................................................................................................................................. 70
8. Appendix B .................................................................................................................................. 76
9. Appendix C .................................................................................................................................. 78
10. Appendix D .................................................................................................................................. 82

1. SUMMARY
The objective of this study was to evaluate the capability of the Hamburg Wheel
Tracking (HWT) devices available in the US market and to identify potential issues with
different aspects of AASHTO T 324 standard procedure in order to ensure proper testing and
accurate, reproducible results. Based on the results of this study, researchers were tasked to
provide proposed revisions with commentary to AASHTO T 324 to enable the use of a
performance type specification for Hamburg test equipment.
A comprehensive experimental program was conducted to evaluate the capability of five
commercially available HWT equipment as well as their ability to accurately measure,
control, and maintain the desired test conditions as specified in AASHTO T 324. The
experimental program concentrated on the following items of the current AASHTO T 324-
14:
• Wheel position waveform, frequency, and maximum speed;
• Impression measurement system;
• Temperature measurement and control system;
• Free circulating water on mounting system;
• Wheel dimensions;
• Wheel loads;
• Specimen and track length; and
• Data collection and reporting.
After performing a comprehensive evaluation of the machines conforming to AASHTO T

324, it is concluded that there are differences between commercially available HWT
machines in the US market. Furthermore, available HWT machines do not meet all the
requirements set forth in AASHTO T 324 including requirements for the waveform, the
temperature range, and the reporting parameters. One reason for some of the observed
differences is due to the ambiguity of the specification and the lack of detailed requirements
for the different aspects of the test method. The following represents a summary of the main
findings of the experimental program.
Waveform: Results of the experimental program showed that two of the four machines
available in the market were able to produce a sinusoidal wave (Vendors B and D).
AASHTO T 324 specifies that the wheel be required to reciprocate over the specimen such
that its position varies sinusoidally over time.
Temperature control system: Since the majority of the HWT machines do not have a
cooling system, obtaining 25°C (77°F) in the bath is dependent on the incoming water
temperature and was not possible when the water temperature was warmer than 25°C. When
testing at 50°C, even though the average temperatures at the end of 30 minutes of
conditioning were within the specification limit of 50 ± 1°C, some locations in the HMA
specimen were not within the specified range. Therefore, a longer pre-conditioning time is
recommended. On the other hand, the upper range of 70°C specified by AASHTO T 324 is
too high and is not encountered in any regions of the US based on the 50% reliability 7-day

average maximum high pavement temperatures computed using the LTPPBind software.
Based on the results of the survey, the highest test temperature used by the states was 56°C.
Deformation measurements: AASHTO T 324 does not currently specify the locations
of the deformation readings or the number of deformation readings. Current specification
has resulted in major discrepancies among manufacturers, as some machines record
deformations at only five locations while others record deformations at 227 locations along
the track length. Results also suggest that the deformation readings are sometimes not being
recorded at the pre-determined locations along the track.
Data collection and reporting: AASHTO T 324 requires five parameters to be collected
and reported to quantify the performance of a mixture to rutting and moisture susceptibility:
number of passes at maximum impression, maximum impression, creep slope, strip slope,
and Stripping Inflection Point (SIP). Upon review of the current requirements detailed in
AASHTO T 324, one may note that not enough specifics are provided to allow for consistent
analysis and reporting of the five aforementioned performance indicators.
At least seven methods, developed by four manufacturers and two state DOTs, were
identified for analyzing HWT test data and reporting the performance parameters. Two
mixtures (Mix 1 and Mix 2), which were tested using the HWT manufactured by Vendor A,
were selected for analysis by the various methods. Mix 1 was a poor performing mixture that
stripped during testing and Mix 2 was a good performing mixture that did not strip during
testing. For Mix 1, substantial differences were observed amongst the different analysis
methods especially in the reporting of the SIP. Furthermore, some of the available methods
do not report the five performance parameters specified by AASHTO T 324. For Mix 2, only
two of the seven methods successfully identified this mix as a non-stripping mix. In addition
to these discrepancies, the approach adopted by Iowa DOT can only analyze HWT results
obtained from the machine manufactured by Vendor A.
Based on the results of the experimental program, revisions to AASHTO T 324 and to the
configurations of the available HWT machines are recommended. Modifications are
proposed to address equipment capabilities, components, or design features in order to ensure
proper testing and accurate, reproducible results. Proposed modifications are discussed in
this report to ensure repeatable measurements and that the results from different
manufacturers are comparable. These modifications include change to temperature
measurement and range, impression measurement system, data collection, and data analysis
and reporting. In addition to the proposed modifications to the AASHTO T 324
specifications, the vendors may need to modify their equipment to meet the new specification
requirements.
After addressing the proposed modifications to the equipment configurations and to the
specifications, a laboratory experimental program shall be conducted in order to compare the
results obtained with HWT devices from various manufacturers when testing the same
asphalt mixture. The experimental program recommends testing contrasting asphalt mixtures
using the four main types of Hamburg test equipment available in the US market and
comparing the results statistically in accordance with ASTM E 1169 Standard Practice for
Conducting Ruggedness Tests.

2. INTRODUCTION AND RESEARCH APPROACH

This report presents the results of NCHRP Project 20-07/Task 361, Hamburg Wheel-
Track Test Equipment Requirements and Improvements to AASHTO T324. This chapter
describes the problem statement, objective, and research approach.
2.1. Problem Statement
The Loaded Wheel Test (LWT) is a laboratory-controlled rut depth test that uses loaded
wheel(s) to apply a moving load on hot-mix and warm-mix asphalt (HMA, WMA) specimens
to simulate traffic load applied on asphalt pavements. In the 1970s, Helmut-Wind
Incorporated of Hamburg proposed a test method and developed specification requirements
to measure the combined effects of rutting and stripping susceptibility. The equipment
developed was named the Hamburg Wheel Tracking Device (HWTD) and has been used for
over four decades worldwide. The HWTD measures the combined effects of rutting and
moisture damage (stripping) by rolling a steel wheel across the surface of an asphalt concrete
slab that is immersed in a temperature-controlled water bath. The interest and use of LWT in
performance specifications, alternatively referred to as rut testers or torture testers, has seen
an increase in recent years. This interest can be attributed to several factors, including the
use of such devices by FHWA and many state Departments of Transportation (DOTs). Other
important factors in this increased popularity are the ease of use and good correlation to field
performance, which led many DOTs to incorporate LWT tests in their specifications as a
pass or fail acceptance criteria.
As the popularity of this test equipment increased, several manufacturers started
producing their own variation of the LWT, while others adapted their existing designs from a
load over a rubber hose to deadweight loading from a steel wheel. Those machines were built
using various solutions for controlling the wheel speed, measuring the rut depth, water bath
temperature control, and reciprocating mechanisms, to name a few. These different machines
are all currently being used by highway agencies and research centers. Despite the
aforementioned discrepancies among the different LWT machines, no comprehensive study
has been conducted to compare the results from different manufacturers.
In 2010, Shiwakoti et al. carried out a research study focused on wheel tracking devices
to develop a rapid test method to evaluate moisture sensitivity (1). The Asphalt Pavement
Analyzer (APA) and the Hamburg Wheel Tracking Device (HWTD) were used for this
research. Compacted cylindrical samples were fabricated using the Superpave Gyratory
Compactor. However, the APA tests were carried out using the rubber hose instead of the
metal wheel. Results showed major differences on the stripping behavior. APA results did
not indicate any stripping inflection points, contrary to the HWTD results that showed
significant stripping susceptibility. A recent study carried out by the Iowa DOT (2)
statistically evaluated the results from 150 test runs on gyratory specimens using a two‐wheel
HWTD manufactured by Precision Machine and Welding (PMW). Linear variable
differential transducers (LVDTs) were used to measure rut depths at eleven locations across
the wheel track per pass. Measurements were recorded to the nearest 0.01 mm every 20th
pass for the first 1,000 passes. The frequency was reduced to every 50th pass thereafter.
Results indicated that the impression measurement location was found to be a source of
significant variation in the HWTD. The study suggests that the differences are likely due to

the non-uniform wheel speed across the specimen, geometry of the specimen, and air void
profile.
2.2. Research Objective
The objectives of this research as stated in the project description are to [1] document the
capabilities of available commercial Hamburg test equipment, [2] determine Hamburg test
equipment capabilities, components, or design features that ensure proper testing and
accurate, reproducible results, and [3] provide proposed revisions with commentary to
AASHTO T 324 to enable the use of a performance type specification for Hamburg test
equipment. In this study, reference to AASHTO T 324 implies reference to the latest
standard published in 2014, AASHTO T 324-14.
2.3. Research Approach
The approach to be followed in this research project was consistent with the guidelines
outlined in the project description. The proposed research activities were divided into five
tasks. Task 1 consisted of collecting and critically reviewing all available Hamburg test
equipment capabilities, and specifications. In Task 2, laboratory experiments were
conducted to determine the capabilities of available Hamburg equipment and the adequacy of
AASHTO T 324. In Task 3, and based on the results of Task 2, revisions were proposed to
AASHTO T 324 to ensure repeatability and accuracy of measurements. In Task 4, a
statistically based experimental plan was developed to validate proposed requirements for
Hamburg equipment and for specimen preparations and their impacts on test results and
acceptance test criteria. Finally, Task 5 consisted of preparing a final report that summarizes
the project findings and conclusions, document the study results, and presents recommended
revisions to AASHTO T 324. Detailed descriptions of the proposed research effort are
presented in the following sections.
2.3.1 Task 1 – Available Hamburg Test Equipment Specifications
The objective of this task is to conduct a critical review of the test capabilities,
specifications, and similarities and differences of available Hamburg test equipment in the
US. AASHTO T 324 “Hamburg Wheel-Track Testing of Compacted Hot Mix Asphalt
(HMA)” establishes the testing protocol as well as the data-reporting format. However, the
standard in its most current version vaguely describes some critical aspects of the testing
procedure and data analysis, leaving room for ambiguous interpretation. Further, some
commercially available machines do not fully comply with its equipment requirements
resulting in discrepancies in the reported results. A comparative review of some of the
critical technical aspects of the representative equipment in the US was carried out and
presented to what is required by AASTHO T 324. It is worth noting that results of this task,
which are presented subsequently in this report, identified four major manufacturers of HWT
in the US. These vendors are referred to as vendors A, B, C, and D to protect the anonymity
of the vendors.
2.3.2 Task 2 – Engineering Desk Analysis of Existing Hamburg Test Systems
In this task, engineering desk analysis was conducted to identify potential issues on
different aspects of the AASHTO T 324 procedure, mainly on its specifications of what

needs to be measured, and the needed accuracy and resolution of the measurements. As
these critical points are identified, the work progressed to evaluate the capability of the
existing equipment to accurately measure, control, and maintain the desired test conditions.
Finally, the minimum equipment capabilities, components, and design features to ensure the
consistency and accuracy of the test were presented. This task concentrated on the following
items of the current AASHTO T 324:
• Loading mechanisms;
• Specimen dimensions; and
Other factors within the current standard were also analyzed to accommodate any
changes and new recommendations as needed. The current issues on each of these sections,
as well as potential improvements were evaluated and presented.
2.3.3 Task 3 – Propose Revisions to AASHTO T 324
Based on the results of Task 2, revisions to AASHTO T 324 were proposed to
incorporate the equipment capabilities, components, or design features that ensure proper
testing and accurate, reproducible results. Modifications were based on the aforementioned
components and whether existing HWT equipment possess the needed technologies to meet
the required specifications.
2.3.4 Task 4 – A Framework for Future Laboratory Evaluation
Upon completion of Tasks 3, Hamburg test equipment capabilities, components, and
specifications would have been reviewed and modified to ensure proper testing, accurate, and
reproducible results such that it may be used in performance-based specifications. In Task 4,
a detailed experimental plan was developed to validate the proposed equipment
configurations and specifications developed in Tasks 2 and 3 and to meet the recommended
modifications to AASHTO T 324.
2.3.5 Task 5 – Prepare and Submit Final Report
The objective of Task 5 was to complete a final report documenting the entire research
effort. Task 5 was divided into two subtasks that include preparation of a draft and a final
report. The report summarized the findings and conclusions and presented the recommended
modifications to AASHTO T 324 along with the developed research framework. Task 5 also
included a detailed review of the capabilities, specifications, similarities, and differences
among the available Hamburg test equipment in the US.

3. METHODOLOGY AND EXPERIMENTAL PROGRAM

3.1. Review of Test Equipment Specifications
In preparation for the upcoming tasks, a critical review was conducted of the test capabilities,
specifications, and similarities and differences of available Hamburg test equipment in the US.
A comparative review of some of the critical technical aspects of the representative equipment in
the US was carried out and presented to what is required by AASTHO T 324. It is worth noting
that results of this task, which are presented subsequently in this report, identified four major
manufacturers of HWT in the US. These vendors are referred to as vendors A, B, C, and D to
protect their anonymity.
3.2. Nationwide Survey
A nationwide survey was conducted to collect information from state agencies on the use of
HWTs. The survey was posted online and was distributed through various LISTSERVs; it was
also announced at related TRB committees. The research team complemented states’ responses
with a review of state specifications available online as well as through email communications,
which allowed a 100% response rate. A copy of the survey is provided as well as the contact
information of survey respondents are presented in Appendix A. The prepared survey consisted
of 13 questions, which are listed below:
1. What type of LWT do you use? (Please choose one or more manufacturers)
2. Does your machine have a single wheel or two wheels?
3. Which specification do you use? (Please choose one)
4. How often do you calibrate your LWT (months)?
5. What does the calibration include?
6. Is your laboratory AMRL certified for AASHTO T-324?
7. What test temperature(s) do you use? (°C)
8. What is the acceptance criteria used in your state? Please attach a copy of your specifications.
9. What type of specimens do you use?
10. Does you agency specify requirements for the Hamburg test specimen fabrication?
11. Do you have test data that you can share? (Please choose one)
12. How is the result of the Hamburg test reported?
13. How do you use the data you obtain from the machine?
3.3. Experimental Program
The objective of the experimental program was to identify potential issues with different
aspects of AASHTO T 324 standard procedure, mainly on its specifications of what needs to be
measured, and the needed accuracy and resolution of the measurements. As these critical points
are identified, the research team evaluated the capability of the existing equipment to accurately
measure, control, and maintain the desired test conditions. Finally, the minimum equipment
capabilities, components, and design features to ensure the consistency and accuracy of the test
were presented. The experimental program concentrated on the following items of the current
AASHTO T 324:
• Wheel position waveform, frequency, and maximum speed;


• Wheel dimensions;
• Wheel loads;
• Specimen and track length; and
Other factors within the current standard were also analyzed to accommodate any changes
and new recommendations as needed. The following sections present the experimental program
conducted to evaluate the aforementioned factors for the different Hamburg equipment available
in the US market.
3.3.1 Wheel Position Waveform, Frequency, and Maximum Speed
Section 5.1 of AASHTO T 324 specifies the movement of the wheel over the specimen. The
wheel is required to reciprocate over the specimen such that its position varies sinusoidally over
time. The frequency of this movement is specified to be 52 ± 2 passes per minute. Additionally,
the maximum speed is specified to be 0.305 m/s (1 ft/s) and is expected to be reached at the
midpoint of the specimen. An extensive evaluation of the HWTs identified in the project was
undertaken to assess compliance with the specifications of section 5.1. HWTs from Vendors A,
B, C, and D were evaluated in this study.
Two approaches were considered to record the position of the HWT wheel as a function of
time. The first approach studied the feasibility of using an accelerometer to measure acceleration
of the sliding mechanism. The acceleration could subsequently be integrated with respect to
time to obtain velocity and integrated once more to yield distance or position. However, this
approach required acquisition of correctly sized accelerometers and signal conditioning
equipment. The second approach studied the possibility of using a video camera to capture
images at a high rate and performing image analysis to obtain the position of the wheel as a
function of time. The second approach was selected as the equipment and accessories needed to
perform the experiment were available in-house.
A GoPro camera was used to capture the video of the HWT wheel during its travel. This
camera was attached to the moving loading arm using an adhesive mount. Figure 1 shows the
camera set up with an adhesive mount. Aluminum slab specimens were fabricated and used to
minimize vibrations during video recording. A ruler was affixed to the top of the slab and the
camera was focused on the ruler. As the loading arm moves along its track, the attached camera
focuses on different parts of the ruler along the slab specimen. The ruler reading coinciding with
the center of each frame of video was recorded during post-processing. This information was
combined with time data (obtained from a video recording rate of 240 frames/second) to obtain a
distance versus time graph.
Two types of rulers were evaluated by the research team and are presented in Appendix B.
Various camera mounting systems (gooseneck/clamp and adhesive mount), camera-to-specimen
distances, and lighting sources were evaluated to obtain an accurate video. The best video
quality was obtained with a non-reflective paper ruler (1/16 in. subdivision), an adhesive mount,
a focus distance of 5 in., and a professional lighting source (Lowel DP).

(a) Overall setup
(b) Camera and ruler

Figure 1
Experimental setup for wheel position analysis
The HWT was allowed to reciprocate for a few cycles before triggering the GoPro camera to
capture video at 240 frames/second. The GoPro camera setup and control were achieved using
the GoPro app on the iPhone. The video data file was further processed as follows:
10

1. The video file was split into individual image frames. Each picture frame obtained was
1280 pixels wide and 720 pixels high.
2. MATLAB software was used to add a vertical red line in the middle of each frame (i.e., to
change the color of column number 640 to red).
3. The images were re-assembled back to a video file.
4. The video was analyzed frame-by-frame and the position on the ruler coinciding with the
red line was noted. The corresponding frame number was also recorded. It should be noted
that the time increment from one frame to the next is 1/240 second.
Figure 2 presents the image of a frame after the red line addition in MATLAB software. In
this frame, the red line coincides with the 2.75 in. mark on the upper scale of the ruler.
Figure 2
Image frame after MATLAB processing
Figure 3 shows a typical plot of the recorded ruler readings or the wheel position as a
function of time, obtained with the aforementioned post-processing procedure.
11

Figure 3
Wheel position as a function of time
3.3.2 Impression Measurement System

Section 5.3 of the AASHTO T 324 specification requires that an LVDT be used to measure
the impression of the wheel as it tracks over the specimen. It further specifies that the LVDT
have a minimum range of 20 mm (0.8 in.) with an accuracy of 0.15 mm (0.006 in.).
Additionally, this system should be capable of measuring the impression at least every 400
passes, recording the number of passes applied, and be able to collect the measurements without
stopping the test.
The current equipment verification procedure detailed in the Appendix of AASHTO T 324
requires that the LVDT be checked in accordance with the applicable ASTM D 6027 procedure
or per manufacturer’s recommendations. However, it does not require verification that the
measurements be recorded at specific locations along the track. These locations, currently set by
the vendors, should be standardized to enable test result comparisons between different vendors.
As a first step, the calibration of the LVDTs was verified for all the HWTs evaluated.
Additionally, an aluminum specimen with a curvature mimicking a rutted specimen was
designed and fabricated to enable verification that the impression readings were being recorded
at the locations specified by the vendors. Figure 4 presents the picture of the fabricated specimen
and the engineering drawing of the specimen is presented in Appendix C. Since the curvature or
“rut” of this specimen is machined per the drawing in the appendix, the depression at any
location along the track is precisely known. The maximum depression of the manufactured
specimen is 19.05 mm (0.75 in.) and is located at the midpoint of the track. The aluminum
specimen allows for verification of LVDT readings and confirms if the readings are being
recorded at the locations specified by the vendors.
12

Figure 4
Metal specimen for verifying locations of deformation readings
During the course of the study, the research team fabricated a new metal specimen with a
longer curved track length to avoid the problem of the wheel “climbing out” of the track. The
machine drawing of this metal specimen and the analytical solution of the wheel and metal-
specimen interaction is presented in Appendix C.
Figure 5
Modified metal specimen for verifying locations of deformation readings
13

The calibration of the impression measurement systems from the vendors were verified as
described in the instruction manuals provided with the machines. Next, the aluminum specimen
was installed in each of the machines to verify that the readings were being recorded at exactly
the locations specified by the vendors. The steps of this procedure are described as follows:
1. The aluminum specimen was flipped upside down to enable the machine to obtain the
“zero” readings. Figure 6(a) presents the picture of the flipped specimen, showing the flat
surface for the “zero” readings.
2. The HWT was allowed to reciprocate for 80 cycles to enable the machine to record “zero”
readings.
3. The aluminum specimen was flipped again to allow the wheel to track over the curved
machine surface. This is shown in Figure 6(b). The aluminum specimen was also centered
along the track of the wheel.
4. The HWT was allowed to reciprocate for 80 additional cycles to ensure that readings of the
curved surface of the aluminum specimen were recorded.
(a) Flat surface for obtaining zero readings
14

(b) Recording deformation readings along the curvature (ruler shown for scale)
Figure 6
Procedure for verifying locations of deformation readings
It is also noted that the impression of the curvature of the metal specimen was recorded by
connecting the electrical output of the machine LVDT to a data acquisition system. As shown in
Figure 7, data were collected at a frequency of 100 Hz. Impression measurement system
readings obtained from all the machines were compared to this reference profile.
15

Figure 7
Curvature of the metal specimen recorded by machine LVDT connected to data acquisition
system
3.3.3 Temperature Control System

The temperature control system comprises a water tank, heater(s), temperature sensor(s),
circulating pump, and controller. AASHTO T 324 specifies that the temperature control system
in the HWT be capable of maintaining the set temperature in the water tank to within ±1.0°C
over a range of 25 to 70°C (77 to 158°F). Additionally, it specifies that the water be
mechanically circulated in the tank to reduce the temperature gradient. The following section
describes the experimental setup used for this purpose.
The current AASHTO T-324 specification requires verification of the temperature in the bath
at four locations. The procedure also specifies the preconditioning time for temperature
stabilization to be 30 minutes. In order to quantify the temperature gradient across the specimen
and to verify that the preconditioning duration is adequate, instrumented hot mix asphalt (HMA)
specimens were used. Four Resistance Temperature Detectors (RTDs) were used with each SGC
specimen, with two RTDs on top and two on the bottom of the slab specimen. Figure 8 shows
the locations of the RTDs in the SGC specimen. Details of the specimen preparation and
instrumentation are presented in Appendix D. The RTDs were connected to a DATAQ DI-
718Bx data acquisition system for monitoring. The data was collected at a sampling rate of 1/8
(or 0.125) Hz.
16

Figure 8
Temperature sensor locations
3.3.4 Wheel Dimensions

The thickness and diameter of the wheels from the different manufacturers were measured. A
digital caliper was used for this task. The measurements were taken diametrically and along the
thickness of the wheels at four different locations as indicated in Figure 9.
Figure 9
Measuring details: The geometry of the steel wheel (Vendor A)
17

3.3.5 Wheel Loads

Section 5.1 of AASHTO T-324 specifies the load on the wheel is 703 ± 4.5 N (158 ± 1.0
lbs.). The load on the wheel is measured by a calibrated load cell. For vendor A, a spacer is
placed to ensure the wheel is horizontally levelled; see Figure 10.
Figure 10
Measuring details: The process of using load cell (Vendor A)
3.3.6 Specimen and Track Length

AASHTO T 324 specifies that two High Density Polyethylene (HDPE) molds be used to
secure the specimen in the testing tray of the machine. The schematic from the specification is
shown in Figure 11. As can be seen, the specification allows some of the dimensions to be
decided by the manufacturer. The wheel track lengths of the evaluated machines were obtained
by analysis of the GoPro data. The ruler reading from the video frames corresponding to the
ends of the track were noted to compute the track length.
18

Figure 11
Specimen mold (reproduced from AASHTO T 324)
3.3.7 Free Circulating Water on Mounting System

Sections 5.5 and 5.6 of AASHTO T-324 requires that the specimen mounting system (slab or
cylinder) must suspend the specimen and provide a minimum of 20 mm (0.8 in.) of free
circulating water on all sides. After filling the water tank and inserting the specimens, the water
depths on all sides were measured using a ruler (Figure 12(a) and (b)). For each tray, six sides of
free water circulating were measured (Figure 12(c)). For each side, three measurements were
conducted and the average results were reported.
19

(a) Measurement procedure (b) Measurement procedure
(c) layout of the measurement location
Figure 12
Free water circulating on the mounting system
3.3.8 Data Collection and Reporting
AASHTO T 324 requires five parameters to be collected and reported to quantify the
performance of a mixture to rutting and moisture susceptibility: number of passes at maximum
impression, maximum impression, creep slope, strip slope, and Stripping Inflection Point (SIP).
In this analysis, the data collection schemes adopted by the vendors were reviewed and
evaluated. Specifically, the number of data points collected and the spacing between the data
points were identified and summarized. In addition, the calculation schemes for the five
performance indicators were reviewed and analyzed. It should be noted that the current
AASHTO T 324 specification only requires data collection at the center (± 1/2 in.) of the track.
However, state agencies utilize different collection schemes in the calculation of the rut depth.
20

4. RESULTS AND ANALYSIS

This chapter presents the main findings of NCHRP Project 20-07/Task 361. It includes the
main findings of the review of the available Hamburg test equipment specifications, the results
of the nationwide survey, and the results of the experimental program.
4.1. Review of Test Equipment Specifications
This section documents the technical specifications of the available Hamburg testers in the
US market. Equipment manufacturers are referred to as vendors A, B, C, and D to protect the
anonymity of the surveyed vendors. The key elements of AASHTO T 324 specifications to
conduct the Hamburg Wheel Track (HWT) test were identified to be the loading mechanism,
temperature measurement and control system, impression measurement system, test specimen
size, and data collection and reporting sections. Tables 1 to 5 summarize the equipment
specifications in terms of load, temperature, deformation, specimen size, and data collection
mechanisms. The relevant sub-sections of AASHTO T 324 are included in each table below.
Table 1. Temperature measurement and control system (AASHTO T 324, section 5.2)
A
Vendor Standard Economy B C D
model model
Type Type T Type T Type J RTD RTD
Room temp
Range (°C) -200 to 350 -200 to 350 0 to 760 -25 to 199
to 70
Number 2 1 1 2 3
Sensor One
between
Next to
Bottom Next to each specimens,
Location each Right side
tank specimen two to be
specimen
positioned
by user
15 (2
Tank volume (gal) 40 18 34 (3 tanks) 22.9
tanks)
Heater 2 x 4.5
Immersion 4.5 4.5 4.0 2 x 1.5
(kW) Heaters
Circulating pump
34 9 11 10 17
(gpm)
Temperature control
tolerance 0.3 0.3 1 1 0.5
(± °C)
21

Table 2. Loading mechanism (AASHTO T 324, section 5.1)

Sinusoidal
Vendor Load Wheel Drive Schematic
Speed
Slider-
A Deadweight No
crank
Pneumatic Scotch-
B Yes
cylinder yoke
Two
indep-
endent
C Deadweight Yes
motors
and
drives
Scotch-
D Deadweight Yes
yoke
22

Table 3. Impression measurement system (AASHTO T 324, section 5.3)

A
model model
Potentiometric
Sensor type LVDT LVDT Magnetostrictive LVDT
position sensor
Range
50.8 50.8 101.6 50.8 50.0
(mm.)
Tolerance
0.15 0.15 0.0762 0.1 0.045
(± mm)
Mounted on Mounted on
Attached to Mounted on side
the side of the side of
Location Top of cylinder the back of of frame in line
the the
loading arm with wheel
specimen specimen
Table 4. Specimen and track length (AASHTO T 324, section 6.4.2)
A
model model
10.671 10.100 10.700 10.700

10.671 inch
inch inch inch inch
9.000 9.000 9.000 9.000 9.060

inch inch inch inch inch
23

Table 5. Data collection and reporting

A
model model
Number of data
Selectable
points collected 11 11 5 227
up to 21
across specimen
Range (± from
4.5 4.5 4.5 4.5 4.45
midpoint), inch
A/D resolution (bit) 16 16 12 17 16
4.1.1 Loading Mechanism

The four HWTs identified apply the load to the specimen using either a dead weight or by
using a pneumatic cylinder. The effect of using these two methods of load application was
investigated as part of the experimental program.
4.1.2 Wheel Speed
The existing HWT devices can be broadly classified into sinusoidal (Vendors B, C, and D)
and non-sinusoidal loading testers (Vendor A). Due to the geometry of the loading mechanism,
the speed of the travelling wheel varies sinusoidally (in the case of Vendors B, C, and D) or in a
non-sinusoidal fashion (Vendor A). The geometry of the non-sinusoidal test machine is shown
in Figure 13. This mechanism is identical to the crankshaft-connecting rod-piston mechanism
used in automobiles and the wheel (crank circle) speed equation can be obtained from
automotive engineering texts.
Figure 13
Crank-slider mechanism
24

There are three consequences of a non-sinusoidal wheel speed: [1] the total time of loading
of the front half of the specimen is less than that of the rear half of the specimen; [2] the average
speed on the front half of the specimen is more than the average speed on the rear half of the
specimen; and [3] the maximum speed is not achieved at the mid-point of the stroke, but rather at
some point on the front specimen.
4.1.3 Temperature Measurement and Control System
AASHTO T 324 test specification currently requires the water bath to be able to control the
temperature from 25 to 70°C (±1.0°C) using a mechanical circulating system. Vendors A and B
use thermocouples while vendors C and D use Resistance Temperature Detectors (RTDs) as
sensors to measure and control bath temperature. Both sensor technologies (thermocouples and
RTDs in conjunction with signal conditioning electronics and analog-to-digital converters) meet
the accuracy requirement of the test method and the text in AASHTO T 324 section 5.2 should
continue to remain technology neutral. The HWTs employ an immersion type heater(s) to heat
and maintain the temperature in the water bath. A pump circulates the water continuously to
minimize the temperature gradient.
4.1.4 Impression Measurement System
In the current version of the AASHTO T 324 test method; there is a requirement of a specific
type of sensor (an LVDT, or Linear Variable Differential Transformer) to measure the rut depth.
The minimum range of this sensor is specified as 20 mm, with an accuracy requirement of 0.15
mm. Table 3 lists the details of the impression measurement systems as implemented by the
evaluated vendors. The sensing technologies used include LVDT, magnetostrictive, and
potentiometric methods. The ranges of these sensors vary between 50 and 100 mm. Table 3 also
shows the locations of these sensors in the various vendors’ designs. Some of the designs
incorporate side-mounts or mounting on the back of the loading arm, while others mount them
on top of the loading arms or the pneumatic loading cylinders. The readings from the various
designs should provide similar results, provided there are no compliance issues.
4.1.5 Specimen Length and Track Length
The cylindrical specimen mounting system in AASHTO T 324 allows some dimensions to be
set by the test system vendor (Figure 2 of T 324). Table 4 shows the total specimen length,
without any gap, for each of the four vendors. The molds provided by vendors A, C, and D
allow a total specimen length of 10.7 in. while the corresponding dimension for vendor B is 10.1
in. Currently, the track length in most of the test machines is close to 9.0 in. These dimensions
were verified as part of the experimental program.
4.1.6 Data Collection and Reporting
Table 5 summarizes the data points collected per cycle and the range of travel covered by the
surveyed vendors. The number of data points collected varies from five (Vendor B) to 227
(Vendor D). These data points can be collected over the entire range of travel of the wheel. For
example, the range of ± 4.5 in. from the center of the specimen equates to a total 9.0 in. track
length that can be covered. The impression measurement requirement in the current version of
AASHTO T 324 specifies that the system be capable of measuring the rut depth “at the center (±
1/2 in.) along the length of the wheel’s path.” However, the report section (10.1) of AASHTO T
324 stipulates that the “maximum impression” be reported.
25

4.2. Findings of the Nationwide Survey
A nationwide survey was conducted to collect information from state agencies on the use of
HWTs. The survey was posted online and distributed through various list serves; it was also
announced at related TRB committees. The research team complemented states’ responses with
a review of state specifications available online as well as through email communications, which
allowed a 100% response rate. References are provided for the following state specifications:
Iowa (2), Montana (3), Colorado (4), Utah (5), Texas (6), Oklahoma (7), Wisconsin (8),
Louisiana (9), California (10), Illinois (11) and Washington (12). A copy of the survey and the
contact information of survey respondents are also provided in Appendix A.
Figure 14 presents the current use of HWTs by the different states. While 21 out of 50 states
indicated that they use HWT (Figure 14), further evaluation of state specifications showed that
only 12 states use it for acceptance of asphalt mixes. Nine states are currently using HWT for
research purposes or are in the process of implementing HWTs in their specifications (e.g., New
Mexico). Of the remaining states, 17 states reported that they use the Asphalt Pavement
Analyzer (APA), which is another type of laboratory wheel-tracking device standardized by
AASHTO T 340 but is not within the scope of this research project.
Figure 14
States Using HWT and APA
The states that currently use HWTs in acceptance of asphalt mixes are California, Colorado,
Illinois, Iowa, Louisiana, Massachusetts, Montana, Oklahoma, Texas, Utah, Washington, and
Wisconsin. It is noted that Wisconsin DOT contracts out HWTs testing but the test is included in
their specifications. It appears from the responses that many states are in the process of adopting
this test in their specifications and one would expect the number of states using the HWT to
increase significantly in the next five years. For instance, Vermont, Georgia, New York, and
South Carolina indicated in their responses the possibility of adopting this test in the near future.
26

4.2.1 Type of HWTs Used

The second question in the survey related to the HWT brands used by the states. Available
vendors are Troxler Electronic Laboratories, Inc.; Pavement Technology, Inc. (PTI); James Cox
& Sons, Inc.; and InstroTek, Inc. Among those 12 states that specify HWT, a Troxler is used;
however, it is noted that the other products are relatively newer and the states may acquire
HWTs from the other vendors in the future. Since HWTs can be conducted on one-wheel or
two-wheel devices, states were asked how many wheels they use. While Massachusetts uses one-
wheel, all other states indicated that they are using two wheels devices.
4.2.2 Test Methods Used
States were polled on the test method used to conduct Hamburg wheel testing in their
specifications. Since AASHTO T 324 does not specify a test temperature, all states need to
modify the test method to reflect local environmental conditions. While Texas, Colorado, and
Montana DOT are using their own state specifications (Tex-242F, CP-L 5112, MT-334,
respectively), all other states use AASHTO T 324 or modified AASHTO T 324 as their
specifications. Kansas and Florida mentioned they use Tex-242F state specifications as
designation if needed.
The survey also polled the respondents on how often their states calibrate their HWT devices.
Based on the responses, seven states indicated that they calibrate the devices every 12 months.
Colorado, Illinois, and Oklahoma indicated that they calibrate the water bath temperature every
six months and all other components every 12 months. Texas and Washington calibrate all HWT
components every six months (Figure 15). It is noted that results presented in Figure 15 are only
for the states that use HWTs for acceptance of asphalt mixes and not for research purposes.
According to AASHTO T 324, water bath temperature needs to be calibrated every six months.
Wheel-load and LVDTs should be calibrated based on manufacturer’s recommendations.
AASHTO T 324 also requires “verifying that the wheel is reciprocating on the test sample at 52
± 2 passes per minute” but does not mention how often it should be calibrated. Based on the
survey responses, all states calibrate load, LVDTs, and temperature. Oklahoma is the only state
that documents a detailed procedure on how to calibrate HWT for temperature, load, LVDTs,
and wheel frequency. From these responses, it appears that HWT calibration needs to be more
detailed in AASHTO T-324 to allow for comparable precision between the states.
The states were also asked whether their laboratory is AMRL certified for AASHTO T 324
or not. Among those 12 states that specify HWT, Colorado, Wisconsin, and Massachusetts are
not AMRL certified for AASHTO T-324. Results were verified from the AMRL website:
http://www.amrl.net/amrlsitefinity/default/aap/r18labs.aspx
27

Load
Temperature
LVDT
Yes
Wheel Frequency No
Horizontal Level
0 2 4 6 8 10 12
Number of Respondents
Figure 15
Calibration of HWT
4.2.3 Test Temperature

Respondents were polled on the water bath temperature. Based on the survey responses and
state specifications, California, Montana, Utah and Colorado use at least two different test
temperatures based on PG, which correspond to 44°C and 56°C. Massachusetts DOT use 45°C
(113°F). The remaining states use 50°C (122°F) only (Figure 16).
Figure 16
Test Temperature(s)
28

4.2.4 Mix Acceptance Criteria

Mix acceptance criteria based on HWT varies among the states. The main results from the
HWT are the maximum rut depth and the stripping inflection point (SIP). Based on the state
specifications, only California, Wisconsin, and Iowa use both maximum rut depth and stripping
inflection point as acceptance criteria (Figure 17). The remaining states use only the maximum
rut depth as acceptance criterion. California specifies the minimum number of passes at SIP for
different PG. Iowa and Wisconsin specify that the ratio of stripping slope to creep slope should
be equal or larger than 2.0. Illinois states that “It may be useful to run every test for 20,000
wheel passes to collect additional data on moisture sensitivity” (11) and Oklahoma states that
SIP may optionally be computed and reported for information. The maximum allowable rut
depth varies among states. Colorado specifies that a maximum rut depth greater than 4 mm (1.57
in) before 10,000 passes be considered a failure. Illinois restricts the minimum number of wheel
passes when maximum rut depth reaches 12.5 mm, which shall be selected based upon the PG
Grades. Among state specifications, California, Colorado, Texas, Louisiana, Iowa, Montana,
Wisconsin, and Illinois clearly list the requirement for the minimum number of passes at a
specific rut depth or the max rut depth at a specific number of passes. Other state specifications
just mention that the criteria shall be selected based upon PG grade or based on specifications but
do not list a specific rut depth. The details of the state specifications are provided in Appendix
A.
Figure 17
Acceptance Criteria Used by the States
States were also polled on the type of specimens. Montana, Utah, and Colorado allow using
either slab or cylinder; other states use cylindrical specimens only given their convenience.
Agencies were also queried whether they specify certain requirements for the test specimen
fabrication. Among the 12 states, only Washington, California, and Louisiana do not specify
requirements for specimen fabrication.
29

4.2.5 Reporting of Test Results

The method adopted to calculate the maximum rut depth varies among state agencies. Since
AASHTO T-324 requires reporting the maximum depression only, seven states report the
maximum rut depth only (California, Colorado, Massachusetts, Utah, Washington, and
Wisconsin). For the other states, the rut depth is calculated by taking the average of several
readings. Texas and Oklahoma use the average of the three centered sensor readings. Illinois
also uses the average of three sensor readings, which include the sensor at which the maximum
rut depth is measured and the two sensor readings around it. Louisiana takes the average of the
five center points. Montana reports the average of the seven center sensor readings. For Iowa, if
the average rut depth at the final pass is larger than 12 mm, they use the average of the five
sensor readings. On the other hand, if the average rut depth is smaller than 12mm, they use the
average of 10 sensor readings.
4.3. Wheel Position Waveform, Frequency, and Maximum Speed
Section 5.1 of AASHTO T 324 specifies the movement of the wheel over the specimen. The
wheel is required to reciprocate over the specimen such that its position varies sinusoidally over
time. The frequency of this movement is specified to be 52 ± 2 passes per minute. Additionally,
the maximum speed is specified to be 0.305 m/s (1 ft./s) and is expected to be reached at the
midpoint of the specimen. An extensive evaluation of the HWTs identified in the project
proposal was undertaken to assess compliance with the specifications of section 5.1. Details of
the experimental program were presented in section 3.
4.3.1 Test Results
4.3.1.1 Wheel Position Analysis
Figure 18(a) presents a plot of the recorded ruler readings or the wheel position as a function
of time for the machine from Vendor A. The fitted curve was plotted using the equation for a
sinusoidal wave. The resulting plot shows differences between the expected and the measured
position readings. This was expected as the HWT from Vendor A is designed to follow the
equation for the slider-crank mechanism and not a sinusoidal wave.
This procedure of recording and processing video data was repeated for the scotch-yoke
mechanism incorporated by Vendor B and the results are presented in Figure 18(b). Results are
compared to the sinusoidal wave. Table 6 presents the equations for the pure sinusoidal and non-
sinusoidal waveforms.
The difference between a pure-sinusoidal machine and a non-sinusoidal machine can be
observed in Figure 18. In the case of a pure-sinusoidal machine, the wheel spends equal amounts
of time on the front and back halves of the track. However, in the case of the non-sinusoidal
machine, the wheel spends more time on the back half of the track (55%) as compared to the
front half (45%).
Figure 18(c) shows the results obtained for the HWT from Vendor C. The wave shape is
characterized by a linear region in the middle (shown as region ‘A’), followed by a slow-down,
and finally by a small stationary duration (shown as region ‘B’) at the track ends. It should be
noted that the wheel travels at a constant rate of speed in the region shown by ‘A’ on the graph.
30

The waveform obtained for the machine from Vendor D is shown in Figure 18(d). This
machine also uses the scotch-yoke mechanism and produces a pure-sine position waveform.
Because of this configuration, the wheel in this machine spends equal amounts of time on the
front and back halves of the track.
31

(a) Vendor A (b) Vendor B (after upgrade to 9-inch track length)
(c) Vendor C (d) Vendor D
Figure 18
Wheel position analysis
32

Table 6. Equations for the position of the wheel

Mechanism
Vendor Slider-crank Scotch-yoke
(non-sinusoidal) (sinusoidal)
2𝜋𝜋𝜋𝜋 2𝜋𝜋𝜋𝜋
A 𝑥𝑥 = 𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟 � � + �𝑙𝑙 2 − 𝑟𝑟 2 𝑠𝑠𝑠𝑠𝑠𝑠2 ( )
𝑇𝑇 𝑇𝑇
B, C, and D 𝑥𝑥 = sin(𝑡𝑡)
where,
x = horizontal position of the wheel (in),
t = time (min),
T = cycle time (=1/26 cycles per minute),
l = length of the connecting rod (in), and
r = radius of the crank circle.
Since LTRC is in possession of two identical machines from Vendor A (referred to A-1, A-
2), the position analysis experiment was performed on both machines to examine the
repeatability of results. Both machines are of the same model type with two-wheel
configuration. Figure 19 presents the results obtained. It can be observed that the curves were
very close to each other, indicating that the results were repeatable.
Figure 19
Wheel position analysis on machines from Vendor A
33

In an effort to quantify the deviations from a pure-sinusoidal waveform, a goodness-of-fit

measure, RMSE (root mean square error), was computed for all the waveforms. The equation
used for computing RMSE is shown in Equation (1) and an example of the computed error is
shown in Figure 20. The results of the computation are shown in Table 7. The pure-sinusoidal
machines from Vendors B and D had the lowest RMSE, followed by the machine from Vendor
C, while the non-sinusoidal machine from vendor A exhibited the highest RMSE values.
∑ 𝑒𝑒𝑖𝑖2
𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅 = � (1)
𝑛𝑛
where,
ei = deviation from a pure sinusoidal curve, and
n = number of data points.
Figure 20
Illustration of Root Mean Square Error computation
The absolute mean deviation (AMD) was calculated according to the following equation:
𝑛𝑛
1
𝐴𝐴𝐴𝐴𝐴𝐴 = �|𝑥𝑥𝑚𝑚𝑚𝑚 − 𝑥𝑥𝑠𝑠𝑠𝑠 | (2)
𝑛𝑛
𝑖𝑖=1
where,
xmi = measured distance along track, and
xmi = theoretical distance along track for a sinusoidal wave.
34

Table 7. AASHTO T 324, section 5.1 parameters

Vendor Vendor Vendor Vendor Vendor Vendor
A-1 A-2 A-3 B C D
Waveform RMSE (mm) 13.21 14.48 13.21 1.02 3.05 1.02
Waveform AMD (mm) 11.43 14.48 13.20 0.88 3.05 1.01
Frequency (passes per
51.8 52 52 51.2 52.1 52.2
minute)
Midpoint
0.33 0.33 0.30 0.27 0.31 0.31
(m/s)
Maximum
0.33 0.33 0.30 0.27 0.31 0.31
(m/s)
Speed
Distance of
maximum
speed location 17.02 8.89 14.22 0.00 0.00 0.51
from midpoint
(mm)
4.3.1.2 Frequency
The number of passes of the wheel over the specimen is specified to be 52 ± 2 passes per
minute. The video of the wheel reciprocating over the specimen was recorded for one minute
and the data were analyzed to compute the frequency of traversal as follows:
52 ∗ 60
𝑓𝑓 = (3)
𝑡𝑡
where,
f = frequency (passes per minute); and
t = time for completion of 52 passes (seconds).
Table 7 presents the results obtained for the machines evaluated. It should be noted that all
the machines performed within the current tolerance of 52 ± 2 passes per minute.
4.3.1.3 Maximum speed and location of maximum speed

The speed of the wheel travel was obtained by computing a moving linear fit of the data.
Figure 21 presents an example of speed computation at the midpoint of the traversal. In this
case, the slope was computed using five data points around the midpoint of the track. Table 7
presents the results for all the machines evaluated. It can be observed that the maximum speed
35

for the machines evaluated was 0.305 m/s (1 ft/s). The current specification in AASHTO T 324
states that the maximum speed be “approximately 1 ft/s” and that it be reached at the midpoint of
the specimen. The locations of the maximum speeds were close to the midpoint of the specimen
for the machines from Vendor B, Vendor C, and Vendor D while for the machine from Vendor
A, it was obtained at a distance of 13.5 mm (0.53 in) on average from the midpoint of the track.
These results are in conformance with the theoretical computations for the sinusoidal (maximum
occurs at midpoint) and non-sinusoidal configurations. The theoretical location for the
maximum velocity for the non-sinusoidal geometry used by Vendor A is 15.5 mm (0.61 in) from
the midpoint of the track. This is due to the property of the slider-crank mechanism used by
Vendor A, where the maximum velocity occurs when the coupling link is tangential to the crank
circle. The numerical analysis involved solving for the derivative of the velocity equation and
equating it to zero as presented in Appendix C.
Figure 21
Speed computation at center of track
4.4. Experiment I – Wheel Dimensions
Results of the measurements for vendors are shown in Table 8. It is noted that the research
team had access to three machines of Vendor A (referred to A-1, A-2, and A-3), which the first
two are standard model machines and the last one is the economy model machine. Therefore, the
same testing protocol was applied to the three machines to assess whether consistent
measurements are obtained for different equipment from the same vendor. As shown in this
table, some of the wheels’ diameters were slightly below the specified diameter (203.2 mm) in
AASHTO T 324 because of normal wear. The width (47.0 mm) was also slightly greater than
the specified width in AASHTO T 324 due to normal wear, as the wheel tends to bulge with
time.
36

Table 8. Wheel dimensions measured from different manufacturers
Vendor A-1 Vendor A-2 Vendor A-3 Vendor B Vendor C Vendor D
Left Right Left Right Left Right Left Right Left Right Left Right
Wheel Wheel Wheel Wheel Wheel Wheel Wheel Wheel Wheel Wheel Wheel Wheel
203.2 203.1 203.1 203.1 203.1 203.2 203.3 203.3 202.9 202.9 203.6 203.0
Diameter 203.2 203.2 203.2 203.2 203.2 203.2 203.4 203.3 203.0 203.0 203.6 203.0
(mm) 203.1 203.2 203.2 203.1 203.1 203.2 203.3 203.2 203.0 203.0 203.0 203.9
203.1 203.1 203.1 203.1 203.2 203.2 203.2 203.2 202.9 203.0 203.0 203.9
47.4 47.5 47.5 47.6 47.1 47.5 47.7 47.7 47.0 47.0 46.4 46.9
Width 47.6 47.6 47.6 47.5 47.2 47.3 47.8 47.8 47.0 47.0 46.8 46.9
(mm) 47.5 47.5 47.5 47.6 47.1 47.2 47.7 47.7 47.0 47.0 46.9 46.9
47.6 47.6 47.4 47.5 47.1 47.1 47.7 47.7 47.0 47.0 46.3 46.6
Diameter
Avg. 203.1 203.1 203.1 203.1 203.1 203.2 203.3 203.3 203.0 203.0 203.3 203.5
(mm)
Width
Avg. 47.5 47.5 47.5 47.6 47.2 47.3 47.7 47.7 47.0 47.0 46.6 46.8
(mm)
37

It is noted that AASHTO T 324 does not currently set a tolerance for the wheel dimensions
and only specifies the averages. This may need to be revised, as the user has currently no
indication on how much the wheel dimensions can deviate from the specified values. Figure 22
and Figure 23 present the average measurements (diameter and width) as well as their deviations
from the specified values in AASHTO T 324.
204.0
203.8 Left Wheel
203.6 Right Wheel
203.4
Diameter (mm)
203.2
203.0
202.8
202.6
202.4
202.2
202.0
A-1 A-2 A-3 B C D
Vendor ID
Figure 22
Wheel diameters for the different manufacturers
50.0
Left Wheel
49.0
Right Wheel
48.0
47.0
Width (mm)
46.0
45.0
44.0
43.0
42.0
41.0
40.0
A-1 A-2 A-3 B C D
Vendor ID
Figure 23
Wheel width for the different manufacturers
4.5. Experiment II – Wheel Loads
The results of the measurement of load wheel are shown in the Figure 24. Both left and right
wheel load were measured. As shown in Figure 24, except for the Vendor B right wheel, all the
test wheel loads were within the 703 ± 4.5 N (158 ± 1 lbs.) as required by AASHTO T 324. It is
38

noted that all vendors use pound as the unit when calibrating the wheel loads. In this case,
Vendor B right wheel load was 157 lbs., which is within the load requirement.
720.0 left wheel

703.0 ± 4.5 N right wheel
710.0
700.0
Wheel Load (N)
690.0
680.0
670.0
660.0
650.0
A-1 A-2 A-3 B C D
Vendor ID
Figure 24
Wheel loads for the different manufacturers
4.6. Experiment III – Free Circulating Water on Mounting System
AASHTO T-324 specifies that the mounting system needs to provide at least 20 mm (0.8 in.)
of free circulating water on all sides. The measurements of each vendor’s free water length are
presented in Table 9. As shown in this table, a number of machines did not meet the
requirements set forth by AASHTO T-324 and requires modifications. It is also noted that
Machine A-3 was a different model than Machines A-1 and A-2.
Table 9. Free circulating water depth

Vendor A-1 Vendor A-2 Vendor A-3 Vendor B Vendor C Vendor D
Top (mm) 38.1 22.3 34.9 17.5 27.3 20.6
Bottom(mm) 108.0 108.0 98.4 22.2 88.6 90.5
Left(mm) 44.5 47.6 6.4 73.0 71.2 71,4
Right(mm) 227.0 227.0 0.0 0.0 108.0 69.9
Front (mm) 257.2 266.7 217.2 98.4 70.62 196.9
Back(mm) 231.8 231.8 101.6 152.4 179.8 82.6
4.7. Experiment IV – Temperature Measurement and Control System
The goals of this experiment were twofold: (1) to determine if the currently specified
preconditioning duration was sufficient and (2) to determine the temperature uniformity in the
bath. To accomplish this, instrumented hot-mix-asphalt cylindrical specimens were used, with
embedded Resistance Temperature Detector (RTD) sensors. The locations of the sensors and the
labelling convention are presented in Appendix D. Eight RTDs were used, two in each
39

cylindrical specimen. Details of the specimen preparation and instrumentation are also presented
in Appendix D.
The temperature evaluations were initially conducted at 50°C (122 °F), the temperature used
by the majority of users of this equipment. Later, at the request of the panel, 25°C (77 °F) and
70°C (158 °F) (the extremes in the current specification) were added to the evaluation.
4.7.1 Evaluation at 50°C

Figure 25 and Figure 26 present the data collected from the embedded temperature sensors
for the machines evaluated. Results are shown for two standard model and the economy model
machine from Vendor A. Table 10 presents the details of the temperature experiment at 50°C.
In this experiment, the data collection system was turned on prior to immersing the specimen in
the water bath (time = 0). The conditioning period time started once the water in the temperature
bath has attained the target test temperature, which is 50°C in this case (Time for the water bath
to reach 50°C shown in Table 10). If the temperature of the incoming water is cooler than the
setpoint, the machine heater(s) heat the water up to target temperature, during which time the
specimens are also heating up, albeit with a slight delay.
The temperatures at the end of 30 and 60 minutes of conditioning are shown in Figure 27 and
Figure 28. The average temperature after 30 min of conditioning is presented in Table 10. As
shown in Table 10, the average temperatures at the end of 30 minutes of conditioning were
within the specification limit of 50 ± 1°C, even though some of the readings (one sensor for
vendor A and four for vendor D’s machine) were slightly below the 49°C limit. At the end of 60
minutes of conditioning, all sensors were within the tolerance limits.
Table 10. Details of the temperature experiment at 50°C

Average
Time for the Time at the end
temperature
Start time of water bath to of the 30 min
Vendor after 30 min of
experiment (min) reach 50°C conditioning
conditioning
(min) (min)
(°C)
A-1 5 10 40 49.3
A-2 7 13 43 49.7
A-3 5 33 63 50.0
B 10 70 100 49.3
C 0 119 149 49.8
D 1 7 37 49.1
40

(a) Vendor A-1 (b) Vendor A-2
(c) Vendor A-3
Figure 25
Temperature versus time graphs at 50°C (machines from Vendor A)
41

(a) Vendor B (b) Vendor C
(c) Vendor D
Figure 26
Temperature versus time graphs at 50°C (Vendors B, C, and D)
42

(a) Vendor A-1 (b) Vendor A-2
(c) Vendor A-3
Figure 27
Temperatures after 30 and 60 minutes of conditioning, 50°C (Vendor A)
43

(a) Vendor B (b) Vendor C
(c) Vendor D
Figure 28
Temperatures after 30 and 60 minutes of conditioning, 50°C (Vendors B, C, and D)
44


Figure 29 presents the results of the temperature evaluation at 25°C. The data acquisition
was turned on prior to filling the bath of the HWTs. The hot water inlets were turned off and
only water from the cold-water faucets was allowed to fill the bath. In the case of machines from
Vendors A and B, the incoming water temperature was 28.3°C. Even after several hours, the
ambient air temperature of the laboratory/trailer was not sufficient to cool the water to 25 ± 1°C.
For machines from Vendors C and D, the incoming water temperatures were 20 and 24°C,
respectively. Therefore, obtaining 25°C in the bath is highly dependent on the incoming water
temperature. The machines from Vendors A and B were located in Louisiana, where
summertime cold-water temperatures frequently exceed 26°C.
Figure 30 presents snapshots of the temperatures after a conditioning interval of 30 minutes.
The machines from Vendors A and B were not able to attain the set target temperature of 25°C
due to high incoming water temperature, while machines from Vendors C and D reached their
target test temperature after 30 minutes. It should be noted that the HWTs from Vendors A, B,
and C do not include a cooling system whereas Vendor D does include a cooling system.
However, the cooling system was not functional at the time of the temperature experiment.
45

(a) Vendor A (b) Vendor B
(c) Vendor C
(d) Vendor D
Figure 29
Temperature versus time graphs for various vendors (25°C)
46

(c) Vendor C (d) Vendor D
Figure 30
Temperatures after 30 minutes of conditioning (25°C)
47


Figure 31presents the data gathered from the sensors in the HMA specimens after they were
placed in HWT baths. The data collection system was switched on prior to immersing the
specimens in the bath. In the case of machines from Vendors A and C, the time required for the
bath to heat up to 70°C depended on the temperature of the incoming hot water. The incoming
water temperatures were 52.4°C and 50.0°C for the machines from Vendors A and C,
respectively. As the water is heated up to test temperature, the specimens also heat up, with a
minor delay.
The machine from Vendor B incorporates two tanks, a lower conditioning tank and an upper
testing tank. The water is heated up to test temperature in the lower tank and is circulated into
the upper tank at the beginning of the test.
Figure 32 shows the specimen temperatures after 30 and 60 minutes of conditioning. At the
end of 30 minutes, the average temperatures were 69.1°C, 68.4°C, 69.2°C, and 69.7°C for the
machines from Vendors A, B, C, and D, respectively. Two of the sensors showed temperatures
less than 69°C for Vendors A and D, while all the sensors for Vendor B were below the lower
temperature limit of 69°C. It should be noted that all except one sensor were within the
allowable tolerance of 70 ± 1°C at the end of 60 minutes of conditioning. In an effort to increase
the water movement in the machine from Vendor D, a small water circulator was added in the
water bath after the 60 minutes conditioning. The results, presented in Figure 33, show that all
the readings were within the 70±1°C specification after the addition of the circulator.
48

(c) Vendor C (a) Vendor D
Figure 31
Time versus temperature graphs at 70°C
49

(d) Vendor D
(c) Vendor C
Figure 32
Temperatures after 30 and 60 minutes of conditioning, 70°C
50

Figure 33
Temperatures at 30 and 60 minutes and after the addition of a water circulator (Vendor D)
4.8. Experiment V – Impression Measurement System
An aluminum specimen with a curvature mimicking a rutted specimen was designed and
fabricated to enable verification that the impression readings were being recorded at the locations
specified by the vendors. Two metal specimens were fabricated during the course of this study.
The first one was used only in the first quarter of the study. The second metal specimen had a
longer curved track length to avoid the problem of the wheel “climbing out” of the track. All
results presented herein were obtained with the second specimen. The machine drawing of this
metal specimen and the analytical solution of the wheel and metal-specimen interaction are
presented in Appendix C. Figure 34 presents a picture of this curved specimen. Since the
curvature or “rut” of this specimen is machined per the drawing in the appendix, the depression
at any location along the track is precisely known. The maximum depression of the
manufactured specimen is 19.05 mm (0.75 in.) and is located at the midpoint of the track. The
aluminum specimen allows for verification of LVDT readings and confirms if the readings are
being recorded at the locations specified by the vendors.
51

Figure 34
Metal specimen for verifying locations of deformation readings
4.8.1 Test Results

The fabricated metal specimen was used to obtain rut measurements from the HWT
machines. The data obtained from the HWT machines were compared to the reference rut
profile of the metal specimen. Figure 35 presents the results of the experiment for HWT
machines from Vendor A. As can be seen, there are significant deviations from the reference
profile, with a marked skew to the right. These results suggest that the deformation readings
from the LVDT are not being recorded at the pre-determined locations along the track. The
locations of these readings as specified by the manufacturer are -114, -91, -69, -46, -23, 0, +23,
+46, +69, +91, and +114 mm with 0 being the midpoint of the track (a total of 11 readings, 22.9
mm (0.9-in.) spacing between readings).
52

Figure 35
Impression measurement system results from HWT machine (Vendor A)
Figure 36 presents the results for the HWT machine from Vendor B. The vendor-specified
locations of the readings are -97, -32, 0, +32, and +99 mm, with zero being the midpoint of the
track. The results show a reasonably good agreement with the expected rut depths.
Figure 36
Impression measurement system results from HWT machine (Vendor B)
53

Figure 37 shows the results of the evaluation for the machine from vendor C. This machine
records data at 23 equally-spaced locations across the track (-110, -100, -90, -80, -70, -60, -50, -
40, -30, -20, -10, 0, +10, +20, +30, +40, +50, +60, +70, +80, +90, +100, and +110 mm). The
data shows good agreement with the metal-profile in the region from -80 to +80 mm. Outside of
this region; the readings seem to deviate slightly from the expected rut-depths.
Figure 37
Impression measurement system results from HWT machine (Vendor C)
Figure 38 shows the results of the evaluation for the machine from vendor D. For this
machine, the readings were spaced 1 mm apart and were taken from -113 to +113 mm along the
track, resulting in 227 readings. The readings are very close to the expected metal-profile, with a
slight deviation towards the right end of the graph. It is possible that the metal specimen was not
completely level with respect to the deformation measuring system.
54

Figure 38
Impression measurement system results of HWT machine (Vendor D)
In an effort to quantify the deviations of the readings from the metal profile, the root mean
square error (RMSE) was computed for measurements from each machine. Table 11 shows the
RMSE values for each of the four vendors. In line with the visual observations, the RMSE
values for Vendors B, C, and D, were lower compared to that for Vendor A, with the lowest
value obtained for Vendor D’s machine.
55

Table 11. RMSE values for impression measurements of metal profile
Vendor RMSE (in.) AMD (in.)
A-1 0.10 0.08
A-2 0.14 0.12
A-3 0.08 0.06
B 0.02 0.01
C 0.02 0.01
D 0.01 0.00
4.9. Data Collection and Reporting
AASHTO T 324 requires the following five parameters to be collected and reported to
quantify the performance of a mix to rutting and moisture susceptibility (stripping):
1. Number of Passes at Maximum Impression: At a fixed maximum impression value (e.g.,
12.5mm), an asphalt mixture with a larger number of passes is more resistant to rutting (13).
2. Maximum Impression: The maximum impression obtained at the completion of the test is
reported to quantify the rutting resistance.
3. Creep Slope: The creep slope is the inverse of the deformation rate in the creep phase. The
creep phase starts after the consolidation phase and ends before the stripping starts. In this
phase, the rut depth starts to increase steadily due to viscous flow. A mixture with a larger
creep slope value is more sensitive to rutting (14).
4. Strip Slope: The strip slope is the inverse of the deformation rate at where the rut depth
increases tremendously as moisture damage occurs. A mixture with a larger strip slope value
is more sensitive to moisture damage. Furthermore, the ratio of the creep slope to the strip
slope has been used to quantify moisture sensitivity in some states (15, 16).
5. Stripping Inflection Point (SIP): The stripping inflection point is usually reported in wheel
passes. This point occurs where the curve has a sudden increase in rut depth and reflects the
phase where the asphalt binder starts to strip from the aggregate. Graphically, the SIP is the
intersection of the creep slope and the strip slope (17, 18).
Upon review of the current requirements detailed in AASHTO T 324, one may note that not
enough specifics are provided to allow for consistent analysis and reporting of the five
aforementioned performance indicators. For example, AASHTO T 324 does not define how to
find the “steady-state portion” to plot the creep slope. At least seven computer programs,
developed by four manufacturers and two state DOTs, were identified for analyzing HWT test
data and reporting the necessary parameters. The methods are briefly discussed below:
56

• In vendor A’s software, the user specifies the locations of the creep and strip regions. In this
approach, the user chooses the “start” and “end” pass numbers for the creep and strip regions
and the software draws straight lines using these points to obtain the creep and strip lines.
• For vendor B, the user specifies a “criterion of change” defined as a given amount of change
over a certain number of passes, e.g., 1 mm over 1,000 passes. The program then computes
the stripping inflection point and subsequently draws the creep and strip lines.
• Vendor C’s program truncates the data to 15 mm rut depth and fits a fourth-degree
polynomial through the rut data. The location of the minimum of the first derivative of the
curve-fit is then determined. A tangential line is drawn at this location to obtain the creep
slope. The maximum value of the first derivative between this point and the end of the data
is used to obtain the strip slope.
• Vendor D’s analysis program involves finding the minimum error from a fitting line. At the
request of Vendor D, details of the approach are not to be presented. However, the results
obtained from this approach will be presented in the next section.
In addition to the HWT manufacturers, several state DOTs and research institute are
developing their own analysis programs to process the data. Iowa DOT has developed a method
to determine the SIP, the details of which are published in Iowa DOT specification “Moisture
Sensitivity Testing of Asphalt Mixture” (2). Oklahoma DOT uses a modification of the Iowa
DOT test method (7). Texas DOT has adopted a new method based on research published by
Yin et al. (19) and uses a program developed by Thunderhead Testing, LLC that implements this
approach. These methods are briefly described below:
• Iowa DOT’s program uses a 6-degree polynomial to fit the rut data. The minimum of the
first derivative of this fitted curve nearest the end of the test is obtained. The tangent line at
that point is the strip slope. The creep slope is located by equating the second derivative to
zero where prior to the strip pass point. It should be noted that the program calculates the
strip slope prior to the creep slope (2).
• The procedure used by Oklahoma DOT is similar to Iowa’s method. However, a sixth-
degree polynomial is used to fit the data. Next, the rut depth at a 1000 passes is determined.
The program then adds 1 mm to this rut depth and finds the number of passes where this
second rut depth occurs. A line drawn through these two points is defined to be the creep
line. To find the strip line, the program determines the minimum value of the first derivative
between 1,000 passes and the end of data. The tangent at this point is defined to be the strip
line (7).
𝑣𝑣𝑣𝑣
• Texas DOT defines three new parameters in its analysis approach: LCSN, LCST, and ∆Ɛ10,000.
The number of passes at which the second derivative is equal to zero is defined to be LCSN.
Then, the rut depth is separated after LCSN to two parts: accumulation of viscoplastic strain
from reciprocate load cycles and from stripping. Viscoplastic strain from loading can be
predicted using Tseng-Lytton model. As a result, the deformation from stripping is the total
rut depth (natural log fitted curve) minus the deformation under loading (Tseng-Lytton
model). In this approach, the number of passes needed to reach the predicted stripping strain
after LCSN is defined as LCST. In Texas DOT, the predicted stripping strain is 12.5mm. The
third parameter is determined by taking the derivative of the projected viscoplastic strain
using the Tseng-Lytton model at 10,000 cycles (19). Since the TTI method of analyzing
moisture sensitivity is not consistent with the performance parameters defined in AASHTO T
324, the results of this approach are not discussed in this report.
57

4.9.1 Test Materials

Two mixes, which were tested using the HWT manufactured by Vendor A, were selected for
analysis by the various methods. While Vendor A recorded rut data at 11 points across the wheel
track, the rut data corresponding to the center reading, or Point Number 6, was used in the
analysis and is presented in Figure 39. As shown in Figure 39, two contrasting mixes were
selected; a poor performing mix that stripped during testing and a good performing mix that did
not strip during testing.
Mix 1 with Stripping Phase Mix 2 without Stripping Phase

0 0
-5 -5
Rut depth (mm)
Rut depth (mm)

-10 -10
-15 -15
-20 -20
-25 -25
0 5000 10000 15000 20000 0 5000 10000 15000 20000
Number of Passes Number of Passes
Figure 39
Rut Depth versus Number of Passes for the Selected Mixes
4.9.2 Results and Analysis

Section 9 of AASHTO T 324 requires the following parameters to be reported: the maximum
impression, the number of passes at the maximum impression, creep slope, strip slope, and the
stripping inflexion point. However, the specification leaves the details of the determination of
these parameters up to the vendor/user. The absence of clear definitions for these parameters
could lead to widely varying results being reported. The following summarizes the results
obtained by processing the selected data sets using the various analysis methods.
The analysis techniques used to obtain SIP location from the HWT data were obtained by
requesting details from each of the four vendors. This involved several email communications
with each of the vendors to get a better understanding of the algorithms used. In each of these
cases, the analysis procedures were independently verified by performing the individual steps of
the procedure. The analysis results for Mix 1 are presented in Figure 40.
58

Vendor A
0
-5
rut depth (mm)

-10
-15
data
-20 creep slope
strip slope
-25
0 5000 10000 15000 20000
number of passes
Vendor C Vendor D
0 0
-5 -5
rut depth (mm)
rut depth (mm)
-10 -10
-15 -15
data data
-20 creep slope creep slope
-20
strip slope strip slope
-25 -25
0 5000 10000 15000 20000 0 5000 10000 15000 20000
number of passes number of passes
Iowa DOT Oklahoma DOT

0 0
-5 -5
rut depth (mm)
rut depth (mm)
-10 -10
-15 -15
data data
-20 creep slope -20 creep slope
-25 -25
0 5000 10000 15000 20000 0 5000 10000 15000 20000
Figure 40
Data Analysis for Mix 1 Based on Different Approaches
59

As shown in Figure 40, for Vendor A, the creep region ranged from 2,000 to 6,000 passes
and the strip region from 11,500 to 12,800 passes. Since these regions are not automatically
determined by the program, the results reported could vary by the user. Vendor C’s program
truncates the data to 15 mm rut depth so the curve is shorter than the rest. Vendor D’s program
also truncates 50% of data set after the SIP based on the user input. As a result, the SIP location
for Vendor D had lower number of passes than the other methods. Iowa DOT and Oklahoma
DOT methods are very similar and the methods for obtaining the strip slope are the same.
Results for the non-stripping mix (Mix 2) are presented in Figure 41. Vendor A requires the
user to identify the creep and strip region. Because the no-stripping data set has no strip region,
the analyze results is not classified here. Vendor C’s program locates the SIP at the end of the
data set. It should be noted that Vendor D clearly identifies the data set with no stripping, as
shown in the graph. In addition, if the ratio of the creep to strip slopes is less than two, the Iowa
DOT program concludes that no stripping occurred, as is shown in Figure 41. ODOT method
identifies the SIP at the beginning of the data set. It should be noted that the Vendor C and
ODOT procedures reported a SIP even though stripping did not occur.
Vendor A
-5
rut depth (mm)
-10
-15
-20
data
-25
0 5000 10000 15000 20000
number of passes
60

Vendor C Vendor D
0 0
-5 -5
No Stripping Phase
rut depth (mm)
rut depth (mm)

-10 -10
-15 -15
data
creep slope
-20 -20
strip slope data
-25 -25
0 5000 10000 15000 20000 0 5000 10000 15000 20000
Iowa DOT Oklahoma DOT

0 0
-5 -5
No Stripping Phase
rut depth (mm)
rut depth (mm)
-10 -10
-15 -15
data data
-20 creep slope -20 creep slope
-25 -25
0 5000 10000 15000 20000 0 5000 10000 15000 20000
Figure 41
Data Analysis for Mix 2 Based on Different Approaches
Table 12 and Table 13 summarize the stripping and no-stripping data sets results from the
different analysis methods. For Mix 1, substantial differences were observed in the reported SIP.
Furthermore, a number of methods could not identify a non-stripping mix such as Mix 2. In
addition to these discrepancies, the approach adopted by Iowa DOT can only analyze HWT
results obtained from the machine manufactured by Vendor A. It is also noted that all programs
provide the rut depth at a 5,000 passes interval. However, two approaches, which are Vendor B
and Oklahoma DOT, did not report the maximum impression and the number of passes at the
maximum impression as required by AASHTO T-324. Furthermore, Iowa program only reports
the rut depth as an average of all 11 data points. For the non-stripping mix, because the
maximum number of passes was 20,000 passes and all seven programs were able to report the rut
depth at every 5,000 passes, number of passes at maximum impression was the same for all
61

seven approaches. AASHTO T 324 also requires reporting the creep slope and the strip slope.
The program provided by Vendors A and B do not report the creep and strip slopes. As shown in
Table 13, only the program provided by Vendor D and the approach adopted by Iowa DOT
successfully identified this mix as a non-stripping mix.
Table 12. Summary of Programs Reporting Parameters (Mix 1)

Number Of Max
Creep Slope Strip Slope
Passes at max Impression SIP
(*10-4) (*10-4)
impression (mm)
Vendor A 12,800 25 N/A N/A 10,712
Vendor B N/A N/A N/A N/A 473
Vendor C 12,800 25 64 34 9,471
Vendor D 12,850 26 8 25 9,104
Iowa DOT 12,806 25 4 53 10,552
Oklahoma DOT N/A N/A 6 107 11,295
Table 13. Summary of Programs Reporting Parameters (Mix 2)

Number Of Max
Creep Slope Strip Slope
Passes at max Impression SIP
(*10-4) (*10-4)
impression (mm)
Vendor A 20,000 2.1 N/A N/A N/A
Vendor B 20,000 2.1 N/A N/A -3,211
Vendor C 20,000 2.1 0.07 0.07 19,892
Vendor D 20,000 2.1 No stripping No stripping No stripping
Iowa DOT 20,000 1.9 0.3 0.4 No stripping
Oklahoma DOT 20,000 2.1 1 3 180
Table 14 compares the seven approaches in terms of the reporting parameters that are
required by AASHTO T 324. As shown in Table 14, Vendor C, Vendor D, and Iowa DOT
report all five indices, as required by AASHTO T 324. However, Vendor B and Oklahoma DOT
do not report the maximum impression and final passes values. Vendor A and B do not provide
the creep and strip slopes.
62

Table 14. Summary of Seven Programs Reporting Parameters

Max
Number Of
Impression Creep Slope Strip Slope SIP
Passes
(mm)
Vendor A Y Y N N Y
Vendor B N N N N Y
Vendor C Y Y Y Y Y
Vendor D Y Y Y Y Y
Iowa DOT Y Y Y Y Y
Oklahoma DOT N N Y Y Y
Y: The program provides this parameter
N: the program does not provides this parameter
4.10. Proposed Modifications to AASHTO T324 Specifications
Based on the results of the experimental program, revisions to AASHTO T 324 are proposed
to incorporate the equipment capabilities, components, or design features that ensure proper
testing and accurate, reproducible results. The key elements of AASHTO T 324 specifications to
size, and data collection and reporting sections. The following issues need to be addressed in the
current specification:
• Section 5.1: It is proposed to define a tolerance for the wheel dimensions. Based on the
results of the experimental program and assuming an acceptable deviation of 1% around the
mean value, it is recommended to specify a 203.2 ± 2 mm diameter, 47.0 ± 0.5 mm wide
steel wheel. It is noted that wheel dimensions tend to change with wear and deviation from
the recommended specifications will necessitate the replacement of the loading wheel.
• Section 5.1: AASHTO T 324 specifies that the wheel be required to reciprocate over the
specimen such that its position varies sinusoidally over time. Since not all the machines
available in the market are able to produce a perfectly sinusoidal wave, a maximum level of
deviation from a perfectly sinusoidal wave should be specified in AASHTO T 324. Based on
the results of the experimental program, the greatest root-mean square error (RMSE) should
be set at 2.54 mm.
• Section 5.1: AASHTO T 324 does not set a tolerance for the maximum speed of the wheel.
It is recommended to add a tolerance of ± 0.02 m/s.
• Section 5.2: AASHTO T 324 specifies the use of a water bath capable of controlling the
temperature within ±1.0°C over a range of 25 to 70°C (34°F over a range of 77°F to 158°F).
Results of the temperature experiment revealed major shortcomings in this part of the
specification:
63

 Since three of the four machines available on the market do not have a cooling system,
it is virtually impossible to set the target temperature to 25°C, especially during
summer time. It is recommended to modify the low range to 35°C (95°F).
 The upper range of 70°C is too high and is not encountered in any region of the US.
Test temperature is usually selected based on the 50% reliability 7-day average
maximum high pavement temperatures computed using the LTPPBind software (20).
In NCHRP Project 9-29, the highest pavement temperature was calculated based on
LTPPBind to be 58°C for Phoenix, AZ (21). Furthermore and based on the results of
the survey, the highest test temperature used by the states was 56°C. Therefore, the
recommended upper range should be changed to 64°C.
 The 30-min preconditioning time specified in Section 8.9.2 is not sufficient to ensure
that all areas of the test specimen have reached the specified temperature within
±1.0°C. It is recommended to increase the preconditioning time to 45 min.
• Section 5.3: AASHTO T 324 does not currently specify the locations of the deformation
readings or the number of deformation readings. Current specification has resulted in major
discrepancies among manufacturers, as some machines record deformations at only five
locations while others record deformations at 227 locations. Results also suggest that the
deformation readings are sometimes not being recorded at the pre-determined locations along
the track. To this end, two major modifications are recommended:
 Specify that deformation readings should be recorded at 11 locations along the length
of the track. These locations should be set at -114, -91, -69, -46, -23, 0, +23, +46,
+69, +91, + 114 mm with zero being the midpoint of the track. The midpoint of the
track should be marked by the different manufacturers to assist the user. While a
manufacturer may elect to record deformations at more than 11 locations, these
locations should be kept consistent to allow for comparisons between the measured
rut depths among different LWT machines.
 Specify that the locations of the deformation readings should be verified
experimentally using the aluminum apparatus developed in this study and presented
in Figure 33. The maximum total RMSE at the 11 pre-set locations should be set at
1.27 mm.
• Section 9.2: A coherent method of reporting the measured rut depth is needed and is
currently not provided in AASHTO T 324. The availability of a consistent method of
reporting the rut depth would allow for comparisons between the measured rut depths among
different vendors. To this end, it is recommended that the average rut depth be calculated
based on the five middle deformation sensors (i.e., sensors located at -46, -23, 0, + 23, and +
46 mm). This recommendation is similar to the work reported by Schram, Williams, and
Buss (22). This study suggested reporting the average measurements of locations 5 through
9 when the average rut depth at the final pass is greater than 12 mm. Results were based on
statistical analysis over 135 test runs on cylinder specimens.
• Section 9.3: The recommended method to calculate the stripping inflection point (SIP) and
other reporting parameters is not clearly defined in the current specification. Furthermore, it
may result in discrepancies in calculating this parameter. It is recommended that an
approach similar to the one adopted by Iowa DOT be implemented in the revised AASHTO
T 324 specifications.
64

4.11. Laboratory Experimental Plan for Validation of Proposed Changes
A laboratory experimental plan was developed to validate the proposed equipment

modifications and specifications proposed in this study. The objective of the proposed
laboratory plan is to evaluate proposed modifications to the Hamburg test equipment as well as
the modified AASHTO T 324 developed in this study. The research team envisions that results
obtained from all Hamburg test equipment should be comparable. Currently, there is no
guarantee that this is the case as the methods of measurement and calculation are different. After
addressing the proposed modifications to the equipment configurations and to the specifications,
a laboratory experimental program shall be conducted to compare statistically the results
obtained with different Hamburg test equipment from various manufacturers when testing the
same asphalt mixture. The experimental program should test contrasting asphalt mixtures using
the four main types of Hamburg test equipment available in the US market. Furthermore, the
selected mixes shall include good and poor performers against rutting. To this end, the following
factors should be considered in the proposed laboratory plan:
• Mixture Characteristics:
 Four contrasting dense-graded asphalt mixture types (small and large NMAS, low and
high binder contents, two binder types [e.g., PG 58-22 and PG 76-22])
• Specimen Types and Sizes:
 Plant-mixed laboratory-compacted (PL) and plant-mixed field-compacted (PF)
specimens;
 Two specimen configurations (core and slab specimens).
• Test Conditions:
 Three test temperatures (35, 50, and 64°C)
• Test Protocols:
 Original and modified AASHTO T 324 specifications
• Equipment Manufacturer:
 Four brands of Hamburg test equipment that conform to the revised specification
Considering that a minimum number of three replicates would be required for each test
condition, it is clear that a complete factorial design for the proposed test matrix will not be
achievable given possible time and financial restrictions. Therefore, a fractional factorial
statistically based design should be developed in order to consider the most important
combinations, which will allow the proposed experimental plan to characterize the accuracy,
repeatability, and proposed test configurations. To minimize variability due to mix preparation,
a single technician shall be fabricating all specimens in the experimental program.
Upon finalization of the specifications for the Hamburg test equipment, the research team
envisions that a protocol that complies with ASTM E1169, Standard Guide for Conducting
Ruggedness Tests, will be used for systematic evaluation of the Hamburg test equipment
available from different manufacturers. The results of the five performance parameters obtained
with different HWT equipment would be compared statistically to assess whether observed
differences are statistically significant.
65

5. CONCLUSIONS AND RECOMMENDATIONS

After performing a comprehensive evaluation of the machines conforming to AASHTO T
324, it is concluded that there are differences between commercially available HWT machines in
the US market. Furthermore, available HWT machines do not meet all the requirements set forth
in AASHTO T 324 including requirements for the waveform, the temperature range, and the
reporting parameters. One should acknowledge, however, that some of the observed differences
are due to the ambiguity of the specification and the lack of detailed requirements for every
aspect of the test method. The following represents a summary of the main shortcomings
identified during the testing program.
Waveform: Results of the experimental program showed that not all the machines available
in the market are able to produce a sinusoidal wave. AASHTO T 324 specifies that the wheel be
required to reciprocate over the specimen such that its position varies sinusoidally over time. In
the case of a pure-sinusoidal machine, the wheel spends equal amounts of time on the front and
back halves of the track. However, in the case of the non-sinusoidal machine, the wheel spends
more time on the back half of the track as compared to the front half.
Temperature control system: AASHTO T 324 specifies the use of a water bath capable of
controlling the temperature within ±1.0°C over a range of 25 to 70°C (34°F over a range of 77°F
to 158°F). Since the majority of the HWT machines do not have a cooling system, obtaining
25°C in the bath is highly dependent on the incoming water temperature and was not possible
when the water temperature was warmer than 25°C. When tested at 50°C, even though the
average temperatures at the end of 30 minutes of conditioning were within the specification limit
of 50 ± 1°C, some locations in the HMA specimen were not within the specified range.
Therefore, a longer pre-conditioning time is deemed necessary. When tested at 70°C, all
machines were able to heat the specimen; however, some locations in the HMA specimen were
not within the required range. The upper range of 70°C is too high and is not encountered in any
region of the US. Based on the results of the survey, the highest test temperature used by the
states was 56°C.
Deformation measurements: AASHTO T 324 does not currently specify the locations of
the deformation readings or the number of deformation readings. Current specification has
resulted in major discrepancies among manufacturers, as some machines record deformations at
only five locations while others record deformations at 227 locations. Results also suggest that
the deformation readings are sometimes not being recorded at the pre-determined locations along
the track. Furthermore, the researchers located the center of wheel travel for each of the
machines before performing the evaluations, it is important that the specimen molds be centered
with respect to the travel of the wheel. Therefore, the vendors should mark the center of travel
on the machines to allow users to line up the molds with that mark.
Data collection and reporting: AASHTO T 324 requires five parameters to be collected and
reported to quantify the performance of a mix to rutting and moisture susceptibility: number of
passes at maximum impression, maximum impression, creep slope, strip slope, and Stripping
Inflection Point (SIP). Upon review of the current requirements detailed in AASHTO T 324, one
may note that not enough specifics are provided to allow for consistent analysis and reporting of
the five aforementioned performance indicators. For example, AASHTO T 324 does not define
how to find the “steady-state portion” to plot the creep slope.
66

At least seven computer programs, developed by four manufacturers and two state DOTs,
were identified for analyzing HWT test data and reporting the necessary parameters. Two mixes,
which were tested using the HWT manufactured by Vendor A, were selected for analysis by the
various methods. Mix 1 was a poor performing mix that stripped during testing and Mix 2 was a
good performing mix that did not strip during testing. For Mix 1, substantial differences were
observed between the different analysis methods especially in the reporting of the SIP.
Furthermore, some of the available methods do not report the five performance parameters
specified by AASHTO T 324. For Mix 2, only two of the seven methods successfully identified
this mix as a non-stripping mix. In addition to these discrepancies, the approach adopted by
Iowa DOT can only analyze HWT results obtained from the machine manufactured by Vendor
A.
5.1. Recommendations
Based on the results of the experimental program, revisions to AASHTO T 324 and to the
configurations of the available HWT machines are necessary. Modifications were proposed to
address equipment capabilities, components, or design features in order to ensure proper testing
and accurate, reproducible results. The key elements of AASHTO T 324 specifications to
size, and data collection and reporting sections. Proposed modifications are discussed in this
report to ensure repeatable measurements and that the results from different manufacturers are
comparable. These modifications include change to temperature measurement and range,
impression measurement system, data collection, and data analysis and reporting. In addition to
the proposed modifications to the AASHTO T 324 specifications, the vendors are expected to
modify their equipment to meet the new requirements.
Based on the findings of the experimental program, it is concluded that there are differences
between commercially available HWT machines in the US market. After addressing the
proposed modifications to the equipment configurations and to the specifications, a laboratory
experimental program shall be conducted in order to compare the results obtained with HWT
devices from various manufacturers when testing the same asphalt mixture. The experimental
program recommended testing a range of contrasting asphalt mixtures using the four main types
of Hamburg test equipment available in the US market and to compare the five performance
parameters statistically according to ASTM E1169, Standard Guide for Conducting Ruggedness
Tests.
67

6. REFERENCES
1. Han, J., and Shiwakoti, H., Development of a rapid test to determine moisture sensitivity of hot mix
asphalt (Superpave) mixtures. Kansas DOT, Topeka, Kansas, 2010.
2. “Moisture Sensitivity Testing of HMA”, I.M. 319, Iowa DOT, Ames, Iowa, 2013.
3. “Method of Test For Hamburg Wheel Track Testing of Compacted Bituminous Mixtures”, MT 334-
110-090211, Montana DOT, Helena, Montana, 2011.
4. “Hamburg Wheel-Track Testing of Compacted Bituminous Mixtures”, CP-L 5112, 8 Colorado DOT,
Denver, Colorado, 2009.
5. “Method of Test for Hamburg Wheel-Track Testing of Compacted HMA”, Section 39-12 1.01D,
Utah DOT, Salt Lake City, Utah, 2011.
6. “Test Procedure for Hamburg Wheel-Tracking Test”, TEX-242-F, Texas DOT, Austin, Texas, 2009.
7. “Method of Test for Hamburg Rut Testing Of Compacted Hot-Mix Asphalt (HMA)”, OHD L-55,
Oklahoma DOT, Oklahoma, August 15, 2014.
8. WisDOT Modified Test Procedure for AASHTO T-324, Wisconsin State Department of
Transportation, Wisconsin, 2014.
9. Louisiana Standard Specifications for Roads and Bridges, Louisiana Department of Transportation
and Development, Baton Rouge, Louisiana, 2015.
10. Modified AASHTO T-324 specification, California Department of Transportation, California, 2014.
11. Modified AASHTO T-324 specification, Illinois Department of Transportation, Illinois, 2014.
12. Standard Specifications for Road, Bridge and Multiple Construction, M41-10, Washington DOT,
2014.
13. Hines, M., “The Hamburg Wheel Tracking Device,” Proceedings of the Twenty-Eight Paving and
Transportation Conference, Civil Engineering Department, University of New Mexico, Albuquerque,
New Mexico, 1991.
14. Solaimanian, M., H. John, T. Maghsoud, and T. Vivek. Test Methods to Predict Moisture Sensitivity
of Hot-Mix Asphalt Pavements. Moisture Sensitivity of Asphalt Pavements: A National Seminar. San
Diego, California, Feb. 4–6, 2003.
15. Mohammad L., Abadie C., Gokmen R., and Puppala A., Mechanistic Evaluation of Hydrated Lime in
Hot-Mix Asphalt Mixtures, Transportation Research Record: Journal of the Transportation Research
Board, No. 1723, TRB, National Research Council, Washington, D.C.,1997.
16. Aschenbrener, T. and G. Currier. Influence of Testing Variables on the Results from the Hamburg
Wheel-Tracking Device. CDOT-DTD-R-93-22. Colorado Department of Transportation, Denver,
1993.
17. Schram, S. A. and R. C. Williams. Ranking of HMA Moisture Sensitivity Tests in Iowa. Presented at
92nd Annual Meeting of the Transportation Research Board, Washington, D.C., 2013.
18. Aschenbrener, T., R. L. Terrel, and R. A. Zamora, “Comparison of the Hamburg Wheel Tracking
Device and Environmental Conditioning System to Pavements of Known Stripping Performance,”
Colorado Department of Transportation, CDOTDTD-R-94-1, 1994.
19. Yin F., Arambula E., Lytton R., Martin A., and Cucalon L., Novel Method for Moisture
Susceptibility and Rutting Evaluation Using Hamburg Wheel Tracking Test, Transportation
Research Record: Journal of the Transportation Research Board, No. 2446, Transportation Research
Board of the National Academies, Washington, D.C., 2014.
20. Federal Highway Administration, LTPPBind Version 2.1 (Software), FHWA, McLean, Virginia, July
1999.
21. Bonaquist, R., NCHRP Report 629: Ruggedness Testing of the Dynamic Modulus and Flow Number
Tests with the Simple Performance Tester. Transportation Research Board of the National Academies,
Washington, D.C., 2008.
68

22. Schram, S., Williams, R., and Buss, A., Reporting Results from the Hamburg Wheel Tracking
Device, Transportation Research Record: Journal of the Transportation Research Board, No. 2446,
Transportation Research Board of the National Academies, Washington, D.C., 2014.
69

7. APPENDIX A
70

71

72

73

Table A-1. List of State Specifications
Minimum Number of Passes

Asphalt Binder Grade
at 0.5 inch Rut Depth*
California DOT
PG 58 10000
PG 64 15000
PG 70 20000
PG 76 or higher 25000
Illinois DOT
PG 58 or lower 5000
PG 64 7500
PG 70 15000
Louisiana DOT
PG 58 12000
PG 64 20000
PG 70 (OGFC) 7500
Montana DOT
Produced Plant Mix
Mix Design
PG 58 10000 15000
PG 64 10000 15000
PG 70 10000 15000
Texas DOT
PG 64 or lower 10000
PG 70 15000
Wisconsin DOT
PG 58 20000
PG 64 15000
PG 70 10000
PG 76 5000
Colorado DOT
maximum rut depth > 4mm before 10,000 passes is
considering a failure
*: the test temperature is based on the tables below
74

Table A-2. Test Temperatures
Asphalt Binder Grade Test temperature (°C)

California DOT
PG 58 45
PG 64 50
PG 70 55
Colorado DOT
PG 58 45
PG 64 50
PG 70 55
PG 76 55
Montana DOT
PG 58 44
PG 64 50
PG 70 56
Utah DOT
PG 58 46
PG 64 50
PG 70 54
*: States Iowa, Illinois, Louisiana, Oklahoma, Texas, Washington and Wisconsin DOTs are using 50°C
for all the tests.
75

8. APPENDIX B
Table B-115. Types of rulers evaluated
Material Metal Paper
Background Black White Black
Width (inch) 1 1/2 1 1
Subdivisions 1/8 1/16 1/32 1/64 1/8 1/16 1/32 1/64 1/8 1/16 1/32 1/64 1/8 1/16
76

GoPro camera (being

attached to arm)
Aluminum
specimen
Ruler
GoPro camera
Aluminum specimen
Ruler
Figure 42
Setups on other machines evaluated for position analysis
77

9. APPENDIX C
The following discussion presents the analytical solution of the wheel and metal-specimen
interaction. Figure 43 shows the drawing of the metal specimen used in this study (curvature
with radius R) with a HWT wheel (with radius r) placed over it at a distance of 𝛾𝛾𝑐𝑐 from the
center. As can be seen from the figure, the wheel will come in contact with the metal specimen
tangentially at the point 𝛾𝛾. Therefore, the rut depth reported by the machine LVDT will be less
than the actual rut in the metal specimen at all points except the center. The following steps
present the mathematical derivation to obtain the difference in rut depth reported by the machine
LVDT and the impression of the metal specimen (𝛼𝛼0 − 𝛼𝛼𝑐𝑐 ). It should be noted that the center of
the curvature of the metal specimen is at (0, R).
1. The equation of the circle with radius R is:

𝑥𝑥 2 + (𝑦𝑦 − 𝑅𝑅)2 = 𝑅𝑅 2 (1)
Therefore,
(𝑦𝑦 − 𝑅𝑅) = ±�𝑅𝑅 2 − 𝑥𝑥 2 (2)
2. Since we are dealing with only the bottom half of the circle
(𝑦𝑦 − 𝑅𝑅) = −�𝑅𝑅 2 − 𝑥𝑥 2 (3)
3. Assume a 𝛾𝛾
𝑦𝑦 = 𝑅𝑅 − �𝑅𝑅 2 − 𝛾𝛾 2 = 𝛼𝛼 (4)
+ 𝛾𝛾
𝑦𝑦 ′ = = 𝛽𝛽 (5)
�𝑅𝑅 2 − 𝛾𝛾 2
4. Use r and 𝛽𝛽 to find 𝛾𝛾𝑐𝑐

𝛾𝛾𝑐𝑐 = 𝛾𝛾 − 𝑟𝑟 × sin(𝑡𝑡𝑡𝑡𝑡𝑡−1 𝛽𝛽) (6)
5. Use R and 𝛾𝛾𝑐𝑐 to find 𝛼𝛼𝑐𝑐

𝛼𝛼𝑐𝑐 = 𝑅𝑅 − �𝑅𝑅 2 − 𝛾𝛾𝑐𝑐2
(7)
78

6. Use r and ( 𝜃𝜃 = 𝑡𝑡𝑡𝑡𝑡𝑡−1 𝛽𝛽 ) to find f

𝑓𝑓 = 𝑟𝑟 × cos(𝜃𝜃) (8)
7. Find 𝛼𝛼𝑅𝑅 and 𝛼𝛼0

𝛼𝛼𝑅𝑅 = 𝛼𝛼 + 𝑓𝑓 (9)
𝛼𝛼0 = 𝛼𝛼𝑅𝑅 − 𝑟𝑟 (10)
Maximum speed location computation for the non-sinusoidal configuration
The position of the wheel in the non-sinusoidal machine is described as follows:
𝑥𝑥 = 𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟(𝜃𝜃) + �𝑙𝑙 2 − 𝑟𝑟 2 𝑠𝑠𝑠𝑠𝑠𝑠2 (𝜃𝜃) (11)

where,
θ = crank angle,
r = radius of the crank circle, and
l = length of the connecting rod.
The speed of the wheel is obtained by taking the derivative of the position and is shown below:
𝑟𝑟 2 sin(𝜃𝜃) cos(𝜃𝜃)
𝑥𝑥 ′ = −𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟(𝜃𝜃) − (12)
�𝑙𝑙 2 − 𝑟𝑟 2 𝑠𝑠𝑠𝑠𝑠𝑠2 (𝜃𝜃)
The maximum value of speed is obtained by taking the derivative of speed and equating it to
zero. i.e.
𝑟𝑟 2 �𝑐𝑐𝑐𝑐𝑐𝑐 2 (𝜃𝜃) − 𝑠𝑠𝑠𝑠𝑠𝑠2 (𝜃𝜃)� 𝑟𝑟 4 𝑠𝑠𝑠𝑠𝑠𝑠2 (𝜃𝜃)𝑐𝑐𝑐𝑐𝑐𝑐 2 (𝜃𝜃)
𝑥𝑥 " = −𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟(𝜃𝜃) − − 3 =0 (13)
�𝑙𝑙 2 − 𝑟𝑟 2 𝑠𝑠𝑠𝑠𝑠𝑠2 (𝜃𝜃) ��𝑙𝑙 2 − 𝑟𝑟 2 𝑠𝑠𝑠𝑠𝑠𝑠2 (𝜃𝜃)�
MATLAB software (MuPAD) was used to numerically solve this equation to obtain θ. The
resulting θ was plugged back into the distance equation to obtain position. The position of the
maximum velocity was thus found to be 0.61 in. from the midpoint of the track. It should be
noted that the values of r and l used were 4.5 and 13.0 in., respectively.
79

Figure 43
Geometry of metal specimen and wheel
Figure 44
Difference between the rut of the metal specimen and the LVDT reading
80

Figure 45
Details of the metal specimen (all dimensions are in inches)
81

10. APPENDIX D
Figure 46
Locations of the embedded RTDs
Table 16. Sensor labelling convention

Sensor Sensor
Side Specimen
position ID
1 Left Front Top LFT
2 Left Front Bottom LFB
3 Left Back Top LBT
4 Left Back Bottom LBB
5 Right Front Top RFT
6 Right Front Bottom RFB
7 Right Back Top RBT
8 Right Back Bottom RBB
82

Specimen preparation for temperature verification

Cylindrical HMA specimens were fabricated in the laboratory and a table saw was used to
cut grooves 0.25-inch wide x 0.25-inch deep, for installation of the RTDs and lead wires. Next,
the RTDs were placed in the grooves and centered with respect to the width of the specimen.
Finally, plumber’s putty was used to seal the grooves and keep the RTDs in place. Each of these
steps is shown in Error! Reference source not found.. It should be noted that through-holes for
the bottom RTDs were drilled at an angle of 45° to avoid sharp bends of the lead wire.
(a) Grooves cut and holes drilled (b) Drill press used for holes for bottom
RTDs
(c) Plumber’s putty used to seal grooves (d) Instrumented specimens in machine
after RTD placement
Figure 47
Instrumented specimen preparation
83

RTD and signal conditioner specifications
The RTDs were purchased after evaluating the temperature range and accuracy requirements.
Model HSRTD (class A) RTDs from Omega Engineering, Inc. were found suitable for this
application. Next, signal conditioners to interface these RTDs with data acquisition equipment
were selected and acquired. The signal conditioners excite and amplify 100-ohm platinum, 4-
wire RTDs that are based on the 0.00385 ohm/ohm/°C curve. The RTDs and the data acquisition
system were calibrated by using a NIST-traceable thermometer and a ±0.01°C bath circulator.
Figure 44 presents the details of the calibration setup. All the RTDs were calibrated to be within
±0.1°C.
84

(b) Neslab model RTE 17 Bath

(a) Top view of RTDs in bath circulator
(c) Overall setup

Figure 48
RTD calibration setup
85

NCHRP - 219 Report - Hamburg Wheel-Track Test Equipment

Uploaded by

Copyright:

Available Formats

NCHRP - 219 Report - Hamburg Wheel-Track Test Equipment

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

NCHRP - 219 Report - Hamburg Wheel-Track Test Equipment

Uploaded by

Copyright:

Available Formats

This PDF is available at http://nap.nationalacademies.

Hamburg Wheel-Track Test Equipment

BUY THIS BOOK

FIND RELATED TITLES SUGGESTED CITATION

Hamburg Wheel-Track Test

Contractor’s Final Report for NCHRP Project 20-07/Task 361

Copyright National Academy of Sciences. All rights reserved.

This work was sponsored by the American Association of State

Authors herein are responsible for the authenticity of their materials

Cooperative Research Programs (CRP) grants permission to reproduce

The opinions and conclusions expressed or implied in this report are

Copyright National Academy of Sciences. All rights reserved.

Learn more about the Transportation Research Board at www.TRB.org.

Copyright National Academy of Sciences. All rights reserved.

Copyright National Academy of Sciences. All rights reserved.

4.2.1 Type of HWTs Used ..................................................................................................... 27

Copyright National Academy of Sciences. All rights reserved.

After performing a comprehensive evaluation of the machines conforming to AASHTO T

Copyright National Academy of Sciences. All rights reserved.

Copyright National Academy of Sciences. All rights reserved.

2. INTRODUCTION AND RESEARCH APPROACH

2.1. Problem Statement

Copyright National Academy of Sciences. All rights reserved.

2.2. Research Objective

2.3. Research Approach

Copyright National Academy of Sciences. All rights reserved.

Copyright National Academy of Sciences. All rights reserved.

3. METHODOLOGY AND EXPERIMENTAL PROGRAM

3.2. Nationwide Survey

3.3. Experimental Program

Copyright National Academy of Sciences. All rights reserved.

• Impression measurement system;

Copyright National Academy of Sciences. All rights reserved.

(a) Overall setup

(b) Camera and ruler

Copyright National Academy of Sciences. All rights reserved.

Copyright National Academy of Sciences. All rights reserved.

3.3.2 Impression Measurement System

Copyright National Academy of Sciences. All rights reserved.

Copyright National Academy of Sciences. All rights reserved.

(a) Flat surface for obtaining zero readings

Copyright National Academy of Sciences. All rights reserved.

Copyright National Academy of Sciences. All rights reserved.

3.3.3 Temperature Control System

Copyright National Academy of Sciences. All rights reserved.

3.3.4 Wheel Dimensions

Copyright National Academy of Sciences. All rights reserved.

3.3.5 Wheel Loads

3.3.6 Specimen and Track Length

Copyright National Academy of Sciences. All rights reserved.

3.3.7 Free Circulating Water on Mounting System

Copyright National Academy of Sciences. All rights reserved.

(a) Measurement procedure (b) Measurement procedure

(c) layout of the measurement location

Copyright National Academy of Sciences. All rights reserved.

4. RESULTS AND ANALYSIS

4.1. Review of Test Equipment Specifications

Copyright National Academy of Sciences. All rights reserved.

Table 2. Loading mechanism (AASHTO T 324, section 5.1)

Copyright National Academy of Sciences. All rights reserved.